Corporate marketing aims to raise brand awareness and increase company revenues. Marketing was traditionally carried out using, inter alia, journals, product promotions and telemarketing. The methods, however, were not effective because they didn’t give customers feedback about their product or service experiences. The popularity of some products is greatly decreasing without a clear reason for product retrieval or decrease. The world ‘s behaviour, with increasing diversification of consumer tastes and preferences, has shifted since the advent of media in 2004. The concept led to social media marketing being introduced to bridge the divide between consumer preferences and current products and services (Benson&Morgan, 2018).
In the early 1970s the social networks started to develop and conceptualize the first newsletter board system (BBS). An online page where items are shared and discussions are conducted can be defined as a newsletter board system. There was a technological limit on sharing speed during the period in which the (BBS) was designed and communication was limited to plain text (Bennett). Nevertheless, this restriction did not impede the use of newsletters in the 1990s. The technology advanced in the late 1990s and incorporated social sites filled with rich media as each user had a page which could have been adapted to reflect their personalities. And social networks grew exponentially in the 2000s as Internet users burst around the world (Blumberg, 2014).
The most important aspect of social media is how a user interacts with other users. Companies, companies and social media organizations have different goals. A company may be on social media, because a competitor is on social media too, and they feel that a specific platform (Hallam) needs to be ahead. Additional companies have important objectives such as improved brand awareness, direct sales of social media or customer service in social media (Johansson, 2014). This suggests that the calculation of investment return is a major problem in social media operators.
Today’s forms of consumer engagement and shopping are increasingly powerful and can be compared with mainstream media forms, where companies become more aware. The e-commerce platform is constantly expanding and remains one of the most important e-commerce developments (Huang & Benyoucef 2013; Rust & Kannan 2016). Furthermore, consumers have expressed their comfort and experience in online shopping and the concepts are becoming increasingly popular, in particular during the thriving economic times. However, Vivek, Beatty und Morgan (2012 ) argue most importantly that social media provides companies with the opportunity to engage with each of their consumers personally to develop stronger connections and, as a result , improve consumer buying behaviours.
The Internet penetration in the United States has peaked by 87,27% in the United States, with internet use increasing to the figure of 293 million users, which accounts for 76 % of the total population, according to Perrin and Anderson (2019). Furthermore, there were about 152.069 Wi-Fi sites available throughout the country. Drake et al . ( 2019) reported that seven in ten Americans had admitted they were avid internet users. This number is expected to gradually increase with more urban areas in the United States. Considering the penetration of the internet and its population, it is enough to say that companies now need to go further than traditional methods of consumer engagement and use social media to take more contemporary measures (Knight 2016). This is because, in most parts of the country, social media is bound to gain ever more users as the web penetrates.
Chapter One: Introduction
1.0 Introduction
The use of social networking platforms for the implementation of a promotional initiative for products or services is social media marketing (Tuten & Solomon, 2017). The above process ensures that people and businesses connect to their audiences to promote their brand, promote sales and create website traffic. The social media marketing process consists of publishing content from these platforms, which attract the customers ‘ attention and can listen to them and get involved (Tuten & Solomon 2017). Individuals and businesses can therefore analyze their results that contribute to the decision-making process. The social media platforms Instagram, YouTube , LinkedIn, Facebook , Twitter and Pinterest are among the most important. The marketing of social media has five key pillars, namely planning and publishing, strategy, audio and communication, analysis, reports and advertising. Companies should ensure that they have an appropriate strategy for social media. As such, companies use various ways to measure return on social media marketing investments. The total value of all social media activities carried out by a corporation is refers to social media ROI (Saravana Kumar & Sugantha Lakshmi, 2012). Assessing social media ROI of a company is important, as it shows the possible impact of social media in the company. It also demonstrates the efficiency of the efforts and resources of a business.
Marketing has grown to new levels with social media growth and continues to changing and taking on new dimensions through different social media platforms. Instagram has become one of the most popular social media platforms in the world and originally a smartphone application. The picture driven platform has become a good way for brands to make marketing, create strong fanbases and gather valuable information.
The social media today is one of the best ways to connect with a brand and make it aware of the brand (Fong and Yazdanifard 2014). The number of users of social media continues to rise, with over 3 billion active users estimated. Smartphones and mobile equipment have facilitated and made it more common to connect to such kinds of social networks (Sharma 2018). social networking is among the most popular activities online. Online consumer behavior was transformed by social media platforms (Kaplan and Haenlein , 2010).
1.1 Background
Social media platforms not only provide the opportunities for users to exchange, participate and share ideas, but also provide an opportunity to look for brand information and to get help with various problems. There has been significant growth in the use of social media and many have access to their mobile telephones (Joseph 2019). The Internet penetration of the world population was 53 percent according to the “We are Social and Hootsuite Global Digital Report 2018” There were 42 percent of active social media users, and 39 percent of active social media users (We Are Social, 2018). These figures continue to grow (Kemp, 2018). Smartphones are the preferred choice for online use according to the same study. Furthermore, people use mobile apps seven times longer than mobile web browsers (Kemp, 2018).
In 2017, the Finnish statistics survey discovered that 63 percent of companies use social media for different purposes, (See fig.1.), and the number is growing. The figure was slightly different depending on the company ‘s industry and social networks were used more significantly than smaller companies. The most widely used aim was to develop images and market products for the company, where 89% of the companies used it. Consumer opinions, assessment or questions received were also common purposes and were used by 58% of companies.(Statistics Finland, 2017).
Figure 1. Social media use 2013-2017 in companies that employ at least ten persons in Finland
Customer relationship management is of paramount importance to the development of business. There have been changes in the manner in which customers can be managed. Initially, products could be developed, and customers could not have a choice but to accept such products. Today , businesses create products that satisfy the needs of a customer. Social media is one of the channels through which customer management can be affected because most people in the world have access to at least one social media platform. These people can actively be engaged by the brand managers to ensure they know the brands and the type of services that are available for them. Information travels faster on social media due to the nature of the networks that exist. Hajli (2015 ) indicates that manufacturers and producers can use the intent to forecast demand and supply. It is possible through the observation of trends on social media. It is in line with the argument that is developed by Cristea and Mocuta (2018), where customer management is required, and it is easy to manage customers on social media platforms to ensure the brand grows effectively in the market.
Customer relationship management is of prime importance for business development. The management of customers has changed. Products can be developed initially and only those products can be accepted by customers. Today companies manufacture products that meet the needs of a customer. Social media is one channel that can affect customer management, because a majority of people around the world have at least one social media platform. Brand managers can involve these people actively to ensure that they know the brands and services available. The nature of existing networks speeds up the transmission of information in social media. Hajli (2015 ) says that manufacturers or producers can use the intention to predict demand and supply. It is possible to do this by following trends in social media. This follows the customer management argument of Cristea and Mocuta (2018) to ensure that social media platforms are successfully expanded in the marketplace, and is easy for customers to manage.
Social networking websites contribute to the growth and marketing of companies by allowing brands to reach their target audiences through advertising, to interact with their customers and to conduct market research (Sajid, 2016). By using social media, companies can reach their international customers in a more affordable way (Porter & Kramer, 2019). The move towards social media involvement will most importantly make social media an indispensable element of their business development plan in order to challenge competition. For different reasons, social media sites attract users. Time passes are the most common cause of the use of social media. Users in this category visit the websites of social media if they are not interested (Pittman & Reich, 2016). Another group of users use social media websites to obtain information about new products and services. Their feeling is, in addition to giving them access to good offerings on products and services, to social media keeps them updated on trends, new products and services (Chua & Chang 2016; Kauret al. 2018). There are various ways in which businesses can leverage social media in consumer participation, subject to these categories. Sites like Facebook have provided a fan page tool, so that companies and brands can be present on Facebook and build an online community. Consumers have the possibility to “like” or “follow” the brand’s page with these brands. In addition to posting or comments in the page, consumers can then view updates from that page on their newsfeed.
In addition, the page is also visible to friends on the platform for one particular user ‘s relationship, so that more people are exposed to the fan pages (Luarn, Lin & Chiu, 2015). On the other hand, Twitter enables brands to develop twitter handles that are followed by consumers. After following the Twitter pages, users will be able to visualize the posts of the brand and interact with them through tweets or retweeting. All social media pages have a similar trend.
1.1.1 Paid Likes vs. Organic Likes
If a person views an ad on one page and also like a page, it is paid for as it happened with an anniversary one day (Smitha, 2013). According to the Facebook page. Social media likes to play an important role in social media marketing because it shows the scope and effectiveness of social media marketing for the individual or business. However, pay likes have different meaning for companies in the contemporary world.
Naturally attained likes are the number of people who see and like content for their users without a paid delivery (Smitha, 2013). The number of individuals shown on posts of a user and social media pages is also included. In addition, the action is described as organic if a user comments a business post on the news feed and the sponsored tag does not appear on the post. Conversely, the number of people who view your content through paid promotions is included in paid payments. Paid loves are typically influenced by the ad targeting options of a user and can also affect organic loves. Organic strategy companies often enhance their reputation and online presence on the market. Companies focusing on paid strategy however gain a strong brand image and a strong market knowledge. However, companies should make organic and paid loves, as they are not able to operate independently. In addition, a well-developed biological strategy improves the online presence and reputation of a company while a paid strategy allows the company to increase its brand image and consumer awareness (Smitha, 2013 ). Despite the long-term implementation of this plan, the management of both aspects of social media ensures greater profitability and maximum investment returns.
1.1.2 Organic and Paid Reach
At present, pay – to – play networks are Facebook and other social media networks. Instagram , Facebook, LinkedIn , and Twitter therefore have algorithm feeds and, instead of the initial chronological order, posts are displayed in consumers based on preference and behaviour (Carter, 2013). Organic loves on the Facebook page of a user can reach only 2 percent and are expected to reach 20 percent below their average payment level. Even if organic reach is low, companies must have a consistent and active social media presence (Carter, 2013). Twitter, Facebook and Instagram are platforms where consumers can visit to receive corporate updates and ask questions about different products and services that the company offers. For example, the Facebook page at Cleveland contains 480, 591, but only 8,124 people have been on the imminent snow posting that had been live on for 24 hours (Vadivu and Neelamalar 2015). However, an advertisement regarding the Cleveland.com Sport Awards paid an additional 12,572 people among the 8,352 people it had initially reached (Vadivu & Neelamalar, 2015). Other people who liked the post were not the fans of the Cleveland.com page, but the paid promotion targeted people who were fans of sports in the Cleveland area. Low organic reach should not stop an organic management, but should focus intensely on paid social media and maintain a strong organic strategy. Second, Facebook ads contain a redirection link to the business page of a company. Thus, even if an ad from an enterprise sends users to a link from Facebook on the website, they are likely to go to the Facebook page of the company to learn more about it (Niciporuc, 2014). However, the consumer curiosity ends here if the page is empty or outdated. Therefore, companies should invest considerably in advertising while keeping up with an up – to – date Facebook page.
1.1.3 How Paid Likes Influence a Company’s Profitability
Paid strategies help companies to generate profits because of the exposure to the target audience which they give a company. Therefore, companies must focus on paid likes as a business driven metric in the current business environment. Companies that have invested heavily in paid strategies generally generate return on their investment and may use tools like Google Analytics, social ROI calculators, and the impact of Hootsuite to measure their profitability (De Cristofaro et al., 2014). Thus a company establishes a strong consumer base by increasing awareness in the existing market and the new target markets. This leads to increased sales and profitability of products. Paying for a company also likes to develop a strong brand image, so consumer loyalty influences. In spite of changing market dynamics, such a company can gain profitability through a steady consumption demand.
Furthermore, paid reach helps companies to build on their strategies and take them to the next level. Regular payment for promotions may not be effective, but companies can concentrate on paying two or three times a week for promotions. This allows companies to minimize their costs and build on their marketing. For example, the budget of paid reach can be limited to $500, as it is crucial for trivial metrics to be exaggerated (De Cristofaro et al., 2014 ). In addition, seasonal and wacky Wednesday posts can be used to ensure the most effective promotion of company marketing. The most significant number of viewers would be responsible for such jobs, and hence help an enterprise to establish a substantial market base. Using the wacky postal strategy of Wednesday would also allow an organization to gain five further views.
1.1.4 How Paid Likes Lead to Return on Investments
In companies focus on social media ads for marketing their products and services, bought likes could play a significant role in terms of the image that the company potrays. As, the rule of thumb in Social Media is “Good product/service= many followers+many likes. The role of paid likes to help a company gain market share by creating product awareness and building a strong brand image is attributed to that. In determining business profitability, market share is crucial; so companies that have achieved a high market share are generally more profitable (Mochon et al. 2017). Market share is often of value because it enables an enterprise to realize economies of scale which allow it to operate in highly competitive markets. Such maintains sales in those markets and increases its performance in the business environment against other markets. A company will also be able to gain market power through market share, which means it will gain more profits than its smaller competitors. As such, the businesses benefit greatly and have a long – term effect on their return on investment.
Return on investment is a performance measure that is used to determine the profitability of a business. The return on investment is calculated by dividing the business benefits and the cost of the investment. Moreover, the return on investments is dependent on both the number of investments required to support a specific volume of sales and net profit on sales (Sen et al., 2018). As such, a business’ negative profit-to-sales ratio would lead to a negative return on investments. Companies should, therefore, significantly invest in paid reach as it will help the company increase its sales in the existing market and new markets; thus, an increase in its return on investments (Sen et al., 2018).
Remarketing involves a company marketing its products and services to consumers who had already visited its website then promoting appropriate subjects and offers based on the context of the consumer’s visit (De Cristofaro et al., 2014). For instance, Facebook Custom Audiences allows Facebook to manage email targeting. As such, it uses consumers uploaded email addresses to present them tailored ads. However, the challenge facing the strategy is the ability to identify the company’s consumers. Such can be mitigated through the establishment of appropriate CRM database and segmentation, thus exporting the list of a relevant target market.
The ability to upload consumer lists into social media advertising accounts allows a company to establish a look-alike consumer (Constantinides, 2014). As such, companies should understand main consumer profiles and then invest in advertisements that appeal to others with similar preferences.
Companies can target audiences on distinct interests and beliefs. For instance, a company selling phone accessories can target social media consumers who have shown interest in phones. As such, marketers who practice targeting based on interest techniques usually realize higher revenue returns and conversion for social ads (Constantinides, 2014).
1.1.6 How Organic Likes Influence Return on Investment
Companies’ return on investments on organic likes has depreciated over the years due to the changes in the users’ behaviors and preferences. Currently, social media platforms are all on algorithmic feeds, meaning posts are only shown to the user depending on their previous preferences and actions instead of the initial chronological order (Crumpton, 2014). As such, organic posts from social media platforms like Facebook page currently reach about 2 percent of a user’s followers, and the number is projected to drop in the future steadily. Furthermore, social media companies have formulated strategies that will help mitigate past metric errors. For instance, the Facebook company is changing the way it presents viewable impressions, and the number of organic likes is projected to reduce with more than 20 percent from its current standing number of users (Crumpton, 2014). Such is attributed to organic reach most being used to enhance awareness on social issues rather than the company’s profitability.
Furthermore, companies that have implemented and significantly invested in the organic strategy enjoy a robust online presence in the online market. As such, the companies have gained a strong market position in the online market. The millennials usually influence the online market; therefore, organic strategies will help companies achieve ROI through the sales of services rather than products to online consumers (Geho & Dangelo, 2012). Furthermore, the organic approach helps companies in improving their reputation. Good reputations attract customers to the company since consumers prefer to buy from companies with a good market reputation. Therefore, organic likes will enable service companies such as hospitals to get returns on their investments as they will gain consumers’ loyalty (Buhalis & Mamalakis, 2015). However, organic reach has been on the decline over the subsequent years, and it is projected to arrive at zero. Such is attributed to brands shifting their focus on paid reach as it is reliable for marketing. Organizations that have implemented organic reach should adopt the paid reach to enable them to gain profits as organic likes are steadily losing viability in influencing a company’s return on investments.
1.1.7 How Organic and Paid Likes Support Key Social Media Marketing Activities
1.1.7.1 Creating Conversation Based on Unifying Thread
Social marketers in a company should formulate appropriate ways to engage the target audience based on a shared passion. For instance, a company can create a hashtag or dovetail with an existing hashtag suppose the content is relevant to the target audience. Using hashtags for a particular conversation enables a company to merge all posts into a searchable thread that will help the audience become part of the conversation (Aichner & Jacob, 2015). Also, hashtags help companies to foster new relationships with consumers based on a shared goal.
1.1.7.2 Using Social Media for Content Marketing
Companies should ensure the content plan sits above all their marketing channels, main stories, and when the stories are played (Constantinides, 2014). As such, organic likes can be employed to increase consumer engagement in a company’s content. For instance, if a ticketing company has a page for sporting events and some of its consumers are rugby fans, it can create a new blog about interviewing the current England Rugby captain and contact the consumers. Such can be done through the consumer’s preferred social media platforms and ask them to comment and share the story. Reaching the consumers through their social media platforms means the content is personal and increases the chance of the consumers reading the content (Buhalis & Mamalakis, 2015). However, a company should be realistic about the number of individuals it contacts in person. Such is attributed to resource implication of the move and can steadily grow as the company gradually understands its impact.
Additionally, paid likes can help in amplifying organic content using social media tools to target existing consumers. For instance, using rugby on Facebook can help the company target consumers who like other leading rugby fan pages (Felix et al., 2017). However, paid likes can be utilized in promoting content assets like important news highlights and reports. Such can be done with a relatively low budget and can rapidly measure the intensification impact.
1.1.7.3 Responding, Listening, and Learning
Organic likes are essential for community management since a company does not need paid likes to listen to what the consumers are saying about them. Conversely, a company can utilize paid likes to emphasize vital messages that it shared in organic likes (Evans, 2010). For instance, if a company is involved in a public issue, then the information it provides to consumers who have contacted the company can be part of the broad educational marketing program with paid likes to outspread the message delivery. As such, listening and learning principles should be part of a company’s organic approach to social media, thus should happen through active involvement in the consumer community.
1.2 Aim of the Study
There are a lot of studies concerning social media; however, there seem to be fewer studies that focus on exploring the relevance of return on engagement as a social media engagement metric. As such, there seems to be a clear need for companies and stakeholders to comprehend whether or not investments in social media can truly be measured and to which extent. Therefore, the study explores the impact of paid against organic content and likes on social media. Thus, aiming to clarify the question that almost every brand has postulated at some point, “Is it wise to buy likes, comments, or other social media features to increase brand awareness?” Also, “what kind of brand posts are effective and can drive engagement?”
While the benefits of using Social media in consumer engagement have been established at face value, there still exists a gap when it comes to measuring the impact or results of engaging with consumers through social media (Aluri, Slevitch & Larzelere, 2015; Gill, Sridhar & Grewal, 2017). This point is where the concept of Return on Engagement (ROE) comes in. There has been comprehensive research on the usage of social media from a business perspective; however, research efforts on return on investment have not fully taken off since measuring the Return of Engagement is a new concept for social media marketing. Besides, ROE is being fronted as the ideal performance metric that could measure positive results of consumer interaction through social media.
1.2.1 Formulas for Calculating Return on Investment in Social Media Marketing
Digital marketing involves a series of steps to ensure that a company can find a mathematical overview of its social media ROI. It also involves tracking metrics, collecting data, and analyzing data to help a company improve its strategy (Felix et al., 2017). One advantage of social media marketing is that it enables the company to get a detailed ROI for its marketing initiatives, unlike the traditional forms of marketing that come without the application of mathematical concepts. Therefore, several mathematical formulas can be used to analyze social media ROIs.
1.2.1.1 Computation of Value of Impressions
There are several ways in which organizations can compute the value of impressions gained in social media. For instance, impression value equals the total of the garnered on social media platforms like Facebook, YouTube, and Twitter (Gilfoil & Jobs, 2012). On YouTube, it refers to the total views that the customers have presented on a particular video posted by a company. The value of impressions also refers to the impressions from all other social media platforms plus the website traffic. The resultant value is then multiplied by the average industry. The average industry refers to the quantity that has to be paid or that which a company pays for its Creative Management Platform campaign (Gould & Nazarian, 2018).
1.2.2 What is Return of Engagement?
Return on engagement is simply the payback on your investment in the engagement process. It is the brand strength gained from a particular brand action, strategy, or product. While return on investment (ROI) is measured mostly in numbers, brands today need something stronger. That’s why (ROE) was developed as a social media metric. Engagement rates were the currency of the social media marketing industry in 2019. Currently, there are six main metrics for measuring the return on engagement.
1. Engagement rate by reach (ERR) = total engagements per post / reach per post * 100
Pros: Reach can be a more accurate measurement than follower count since not all your followers will see all your content. And non-followers may have been exposed to your posts through shares, hashtags, and other means. Cons: Reach can fluctuate for a variety of reasons, making it a different variable to control. A very low reach can lead to a disproportionately high engagement rate, and vice versa, so be sure to keep this in mind.
2. Engagement rate by posts (ER post) = Total engagements on a post / Total follower *100
Pros: While ERR is a better way to gauge interactions based on how many people have seen a post, this formula replaces “reach” with followers, which is generally a more stable metric. Cons: As mentioned, while this may be a more unwavering way to track engagements on posts, it doesn’t necessarily provide the full picture since it doesn’t account for viral reach. And, as follower count goes up, the rate of engagement could drop.
3. Engagement rate by impressions (ER impressions) = Total engagements on a post / Total impression *100 Pros: This formula can be useful if you’re running paid content and need to evaluate the effectiveness based on impressions. Cons: An engagement rate calculated with impressions as the base is bound to be lower than ERR and ER post equations. Like reach, impression figures can also be inconsistent.
4. Daily engagement rate (Daily ER) Daily ER = Total engagements in a day / Total follower *100 Pros: This formula is a good way to gauge how often your followers interact with an account daily, rather than how they interact with a specific post. As a result, it takes engagements on new and old posts into the equation. Cons: There’s a fair amount of room for error with this method. For instance, the formula doesn’t account for the fact that the same follower may engage ten times in a day, versus ten followers engaging once.
5. Engagement rate by views (ER views) = Total engagements on video post / Total video views *100. Pros: If one of your video’s objectives is to generate engagement, this can be a good way to track it. Cons: View tallies often include repeat views from a single user (non-unique views). While that viewer may watch the video multiple times, they may not necessarily engage multiple times.
6. Factored Engagement Rate = Comment-weighted ER = (Total comments x 2) + all other engagements / Reach per post *100. This technique inflates the resulting engagement rate and can be misleading, especially since the use of factored engagement rates is not widespread. For this reason, companies such as Hootsuite does not recommend their use.
1.3 Problem Statement
As managers are becoming more comfortable with even including social networks and blogs as part of the company’s marketing communications, attention has been turned to questions with regards to return on investments of social media marketing (John). Return on engagement is there intertwined with the Return of Investment. Return of Investment is defined as a performance measure that is used in evaluating the efficiency of an investment in comparison to the efficiency of different investments. The ROI is therefore used to calculate the additional profit that has been made as a result of an investment such as marketing campaigns. Therefore, the Return on Investment is an evaluative metric. The evaluative metric enables you to assess the direct profitability of investments. A company has to consider all possible ways that social media may benefit it when calculating the Return on Investment (Miller, 2015-2016). The ROI, therefore, indicates how many euros profit will jump out on a Euro capital investment.
Peculiarly, there’s no standard metric for calculating “Investment Engagement Sale.” Current metrics do not take frequent fluctuation of likes, shares, comments, views, and other inconsistencies in the factor. Social media is a fast-paced environment where users have the opportunity to change their opinions within seconds. As such, it is currently difficult for companies to use current methods and metrics to measure the ROE as current ROE formulas do not consider the above factors. What if there was one formula that companies could use to calculate the investment engagement on the product. What if companies could measure how many products they have sold based on how much investment they have made on social media?
1.3.1 What is the Investment Engagement Sale Figure?
Investment engagement is the end outcome of an investment that aims to engage consumers on social media for a product or service (Jesernik, 2019). The current formulas are more focused on the reaction on social media but not on the entire life cycle. The social media investment should not only start and end with social media engagement; instead, any metric that is used to measure social media campaigns must also incorporate the product and sale cycle. Engagement rate is a formula that measures the amount of interaction social content earns relative to reach or other audience figures, for instance, reactions, comments, and shares. Total engagements typically represent a tally of likes, reactions, comments, shares, views, retweets, and sometimes include clicks, depending on which platform companies are using. The investment engagement sale figure metric aims to bridge the gap in the industry standards for controlling success in social media marketing, especially in the area of influencer marketing.
In 2017, influencer marketing dominated the marketing industry, reaching beyond even distribution in daily and business media. Not everyone is convinced about the “Power of Influencers.” While half of the marketing executives consider Instagrammer, Youtuber, and Blogger as the new brand ambassadors, the other half doubt the measurability of the results. Generally, in the transparency of the costs and the success control, valid industry standards are missing. Many consider, “what metrics are employed, and which are relevant.”
The use of fake followers and bots is to attain more engagement, clicks and conversions are currently seen and displayed as a huge problem in the social media marketing industry. The usage of fake followers and bots has grown. Ninety-four percent (94%) of marketers currently believe that future transparency and authenticity will be the key to the success of the influencer. Nonetheless, because of this manipulation in the social media marketing industry, the FTC guidelines call for more transparent cooperation between influencers and brands and, for example, the labeling of sponsored news in the social networks. According to a report by the search engine, Influencer DB, Instagram currently has more than 558,000 influencers, who have more than 15,000 followers. Eighty-one percent (81%) of all influencers are so-called micro-influencers with 15,000 to 100,000 followers and influencers with more than five million followers contribute to one percent (1%) of all influence users on the net.
The revenue potential for companies that focus on influencer marketing is huge. For example, 22 percent of 18 to 34-year-old clients purchased after seeing an online influencer associated with a particular product. And half of the 25 to 34-year-old users from a survey of 1,200 consumers who viewed influencer content or who communicated with influencers via Facebook, Twitter, YouTube, or Instagram believe that influencer communication is authentic and genuine (Source: Marketingcharts.com, 2018).
In a competitive world, every company wants to be number one or among the top ten. Therefore, it was only a matter of time that companies discovered Social media platforms and influencer marketing for themselves. It’s time-consuming and labor-intensive to build quality reach on social media channels. Those who want to assert themselves on the platforms need a lot of dedication and patience to win a solid number of followers for their appearances. However, many companies want their brand and image to be dominated and presented everywhere. Some of these companies, therefore, utilize services from other companies that specialize in growing social media pages of companies with a slogan of “No patience? No problem! You can finally buy followers”. The network is teeming with offers: 1000 Followers on Instagram are already available for less than ten euros. However, the purchase would be a quick way, but not the right one because such a procedure has all parties involved, i.e., influencers, advertisers, social media platforms, which is disadvantageous in the long term.
When compiling this dissertation topic, the researcher did not find any literature, research, or study that measured the metrics of bought social media services such as likes, comments, and followers and the effects that such services have on a business, a celebrity, or an average person. However, being ineffective in some businesses does not mean its ineffective for other organizations or individuals. Measuring the return of investment in social media marketing requires all areas and entities of social media taken into consideration. It is argued that fake followers sell nothing but an empty promise. And as such, companies are advised to invest in quality over quantity.
However, taking into consideration the latest controversy in the United States Presidential Election with regards to fake news, from a scientific perspective, it can be argued that widespread fake news on various social media platforms played a vital role in influencing voters. Thus, in the election of a President of the most powerful country in the world. Therefore, the research will tackle and find metrics to measure the ROI for social media marketing as it relates to organic and bought services. As such, the researcher will not only explore the relevance of return on investment as a social media engagement metric but also develop a formula that will calculate the ROI for bought social media services.
As such, the study was carried out in incorporation with a Start-up called Epoxa GmbH Co KG located in Germany. During the study, Instagram posts were analyzed. The following research questions were raised:
1) What influence do pay likes and comments have on followers of the company on Instagram?
2) Do organic likes and comments have a greater advantage and overpaid services?
3) Could the ROI of organic and paid likes truly be measured?
4) Are businesses attaining success from Social Media Engagement?
5) Is brand engagement an applicable metric?
6) What is the present indications of Return on Engagement?
7) What are the behavioral results of Return on Engagement?
8) How can Return on Engagement be used as a social media engagement metric?
9) What are the procedures that can be used to quantify Return on Engagement?
10) What will you need to ensure when you are selecting content and engagement interventions?
11) Is Brand engagement a relevant metric?
12) What are the forerunners of Return on Engagement?
13) What are the behavioral outcomes of Return on Engagement?
14) What methods can be used to quantify Return on Engagement?
15) Organic vs. non-organic Social Media
a) Characteristics of Organic Content
i. How can organic content be measured?
ii. Is organic content profitable to businesses?
b) Characteristics of Non-organic content
i. Which businesses that sell non-organic content?
ii. Is non-organic content damaging to companies?
16) Real Influencers vs. Fake Influencers
a) Who are influencers, and are they effective?
i. What is influencer marketing?
ii. How to identify real influencers from fake influencers?
iii. Are current metrics able to effectively measure the ROI in influencer marketing?
17) Reasons why companies or individuals buy Facebook Likes?
a) Where to buy fake likes?
b) What are the benefits behind liked information?
c) Which likes the most important?
18) Pros and cons of purchased fans
a) What is more important than many fans?
19) Is it worth it to buy likes and followers?
20) What other sustainable social media marketing for more success?
Chapter Two: Literature Review
2.0 Introduction
Firms endeavor to increase their sales volumes through marketing as a way of increasing their customer base. The changing market dynamics necessitate firms to adopt new marketing approaches in line with the emerging information and communication technologies through social media marketing platforms. The return on investment measures the financial returns from a venture that further enables firms to determine their profitability or losses over a period. The rise in social media marketing has created an opportunity for firms to increase their presence among online consumers while evaluating their performance through customer feedback across social media platforms (Ray, 2009). Business enterprises are continuously utilizing social media for branding, customer management, and public relations and as a way of reaching out to a particular target group. As such, many firms attempt to measure their return on investment following the application of extensive social media marketing through various social media marketing platforms (Edosomwan et al., 2011). With the increase in the use of social media as an integral aspect of advertising, firms often undertake to evaluate the financial profitability of their venture in determining their progress. Therefore, firms need to assess their social media objectives in line with the potential return on investment in the global marketplace. Some of the social media marketing platforms include; Facebook, Google+, Twitter, LinkedIn, YouTube, through which firms can capitalize on their features to promote and publicize their products.
According to Bianchi and Andrews (2015), many organizational leaders and managers are increasingly utilizing social media marketing approaches to increase their competitive advantage with the view of increasing their return on investment. The researchers sought to understand the manager’s perception of social media while highlighting various approaches to measuring the return on investment. Again, Bianchi and Andrews (2015) assert that social media marketing methods utilize the networking capabilities to enhance their online presence, thereby targeting a particular group of individuals. Through networking capabilities, social media platforms can provide rapid and instant feedback regarding consumer perception and the suitability of the product within the target market. It is also possible to determine the extent of satisfaction by providing polls for consumers to evaluate the product or service. Through the social media marketing sites, firms can create mini polls that allow for image transferability of the utilized product or service within their situations in realizing how the firm benefits with the continued usage of the online media advertising approaches (Deepa & Deshmukh, 2013). Besides, managers within the organizations can evaluate the social media metrics by asking the clients interactive questions regarding their experience by determining their online presence in the company’s platform.
Elsewhere et al. (2012) highlight various ways through which a firm can measure their return on investment through social media marketing. By developing new metrics, it is evident that firms can capitalize on the social media users’ discussion and their impact on potential consumers of a particular commodity by developing the customer influence effect. Through the metric, firms have the potential to determine the extent to which their product impact on other users through sharing messages and information from specific user groups. Thus, in a situation where an online user sends several messages to another individual within a group regarding a commodity, a firm may attempt to realize their influence on others while assessing their levels of satisfaction that result in income generation. Besides, Kumar and Mirchandani (2012) point out the need to measure how best an online user would regularly continue communicating about the product to their peers through the stickiness index metric. In realizing that people can consume related commodities, the metric can provide information regarding how the social media users attach to the commodity within their localities. By utilizing the Hokey Pokey case study, it was apparent that the social media campaigns and marketing technique can significantly increase the return on investment in situations where firms develop a consistent approach and strengthen their customer relations (Kumar & Mirchandani, 2012). Consequently, firms need to embrace the collection of data analytics from online platforms as a way of determining the consumers’ level of satisfaction and their profitability.
Additionally, Hoffman and Fodor (2010), in their quest to understand how social media marketing influences the return on investment, emphasized the increasing diversion from the traditional measures of return on investment. With the increase in online media communications and blogs, it is apparent that several managers are considering the customer motivation towards their consumption pattern of a product. Again, firms can measure the return on investment on social media platforms by determining the number of investments consumers make and devote towards their online presence with the product. Hoffman and Fodor (2010) emphasize the need to assess the product’s review, blog posts, and discussion boards in realizing the profitability of their investments. Although the social media presence may fail to generate increases in the return on investment immediately, the management of firms needs to appreciate the essence of steadily building strong consumer relations that would provide critical focus on their social media online marketing approaches (Öztamur & Karakadılar, 2014). Firms have the potential to generate unique content for their products across various social media platforms. Through the contents and customer attributions, firms can determine the changes in the sales volume for their commodities by inquiring about the consumer level of pleasure, motivation, and approval for the products.
On the other hand, Lloret Romero (2011) expresses the need for organizations to adopt social media marketing approaches in a bid to reduce their cost of marketing operations. The article underscores the need to exploit market research techniques such as interviews and surveys in analyzing the impact of social media methods. In measuring the return on investment on social media marketing, Lloret Romero (2011) emphasizes the need to assess the changes in the consumption patterns between the past consumers and the present online user who join in the latter stages. A firm can determine whether they have an improvement in their investment returns by comparing the two groups of online users. Again, organizations can measure the return on investment by comparing the consumption behavior of an individual before joining the company’s website and after joining the social media platforms (Lloret Romero, 2011). Since online users share honest opinions regarding their experiences, it is possible to discern the extent of profitability from how the new users perceive online experiences. Lloret Romero (2011) also notes that firms can measure their return on investment by determining the level of success of a commodity after availing its presence on social media. Therefore, measurement of the investment return needs to adopt an end process methodology in perceiving the brand reception across the social platforms.
Moreover, Fisher (2009) underscores the increasing role of social media in evaluating the return on investments across many firms. Through the evaluations, marketers have an opportunity of advising the management regarding the need to pursue particular aspects of production that improve consumer potential. Fisher (2009) points out the importance of evaluating social media gains through consumer metrics and standards across social media sites in determining the extent of their sales progress and profitability. Besides, Fisher (2009) provides information on determining the social media applications and widgets in realizing their influence and unique users who search for particular company applications. By appraising the consumer usage patterns, conversation in blogs, and pageviews across the social media sites, it is apparent that a firm can estimate its financial returns and value on a specific investment. Although Fisher (2009) recognizes the influence of marketing strategies in measuring the returns, firms must adopt a strategic approach in which the social media reviews, blogs, and conversations result in sales and profits for the firms. Additionally, measuring the return on investment entails monitoring the efficiency of the new products while seeking consumer feedback regarding their willingness to purchase the commodity in the future. Through the approaches, it would be possible to comprehend the actual and potential financial returns.
In addition, Buhalis and Mamalakis (2015) underscore the relevance of social media marketing in improving the sales volume of a company’s product. Although the research points out a potential weakness in measuring the social media marketing approaches to return on investments, it supports other studies that highlight the necessity to adopt the available strategies towards improving financial success. Again, Buhalis and Mamalakis (2015), in their research, highlight a provision for determining the number of social media visits from online users and realizing how they actualize into sales and improved returns. Besides, comprehending the non- financial metrics is critical in evaluating the return on investment, which forms a better predictive tool for the future as opposed to the evaluations based on accounting systems. Some of the evaluation criteria Buhalis and Mamalakis (2015) suggest, include face to face engagements, customer service engagements, accessibility of the commodity, and brand presence across the global market. Thus, online metrics such as likes and comments generate great value in assessing the level of consumer contentment as a means of evaluating the non-financial returns. The higher the level of satisfaction and fulfillment from the online reviews, the higher the potential of the firm to receive non-financial value on their investment (Kumar et al., 2016). Nonetheless, firms need to ensure that they translate the online reviews and comments in monetary returns in realizing and meeting their financial obligations.
Given the relative importance of social media in improving business prospects, Gilfoil and Jobs (2012) emphasize the need for organizations to understand the rationale for pursuing particular social media marketing methods in increasing their brand presence. The authors note that firms need to measure the returns on investment from social media marketing within the broader context of operations. Business entities need to recognize their internal and external environments to undertake specific aspects of social media marketing that would improve the return on investment (Gilfoil & Jobs, 2012). Through the distinct approaches, a firm can determine how social media advertising techniques go along in promoting their overall income. Gilfoil & Jobs (2012) also express the need to develop social media tools customized for the specific investment in enhancing an understanding of the financial and non-financial returns on their investments.
Furthermore, Kaske et al. (2012) note that for firms to realize increased returns, the social media initiatives need to focus on increasing customer loyalty, enhancing communication, and their experience with the product. In measuring the return on investments from the social media approaches, Kaske et al. (2012) notes that the ROI marketing formula can be utilized in determining the changes in customer expenditures and denoting the prospects of the social media marketing techniques. The formula incorporates the net present value in highlighting the consumer expenditures in calculating the future cash flows for the organization. As such, the return on investment framework attempts to evaluate how consumer experiences obtained through the social media platforms result in higher returns for a firm’s investment.
Digital marketing is all electronic marketing on electronic devices and the internet. The most known forms would be emails, text messages, and advertisements on the internet. However, there are also the companies’ webpages, campaign pages, banners, search engines, social media pages, interactive television commercials, and different competitions (Karjaluoto, 2010 p.14). As the number of people who go online every day increasing (Alexander, 2018) and the technology making it easy with more options to access multiple online platforms, digital marketing is becoming more and more important.
There are many options to do digital marketing, and the key is to find the right way to reach already users and potential consumers. Lovett and Staelin (2016) indicate that brands can use emails while for some social media, there is the possibility to use paid (advertising), earned media (word of mouth and online social media), or owned media (brand website and other owned content). Often the best way is to use many platforms and try to link them together, and this way, reach a wider range of potential consumers or engage more with the already existing ones (Mihm, 2012-2014). Facebook, Twitter, and Pinterest are all social media platforms, and investing in these platforms takes a lot of time, but the cost is low (Mihm, 2012-2014).
Digital marketing brings in greater effectiveness than traditional print marketing by using digital analytics software. Digital marketing allows monitoring of real-time data such as how many have viewed the website, what device they were using, where they came from, among other digital data. As such, companies can concentrate on marketing in the right areas. With digital marketing, it is also possible to identify trends and patterns in the behavior of consumers (Alexander, 2018).
2.1 Planning Digital Marketing
There are many models available to help with the planning of digital marketing. Smart Insights created the RACE, and it is simple, secure, and at the same time, goes through the consumer and the company stages. In figure 2., the upper box represents the company stages and the lower the buyer stages.
Figure 2. Digital Marketing Planning System RACE (Smart Insights, 2015)
The RACE starts from the company stages Plan part, and after this, both company and buyer stages should be considered at the same time because they both influence each other. From the company stages, the Plan box represents creating the branded content, defining the segments, targets, and positioning.
Buyer Stage: Exploration – getting to know what is there
1. Reach – Publishing and promoting content on various platforms. This is the part when companies should raise awareness on search engines, social media pages, websites, and others. The media can be paid, earned, or owned.
Buyer Stage: Decision Making
2. Act – Act on the web site, blog, community, and interactive tools. The companies should be worth finding. Information should be relevant, inspirational, and useful to consumers and potential consumers. This is the part where interaction and action are needed, and the same should be wanted from the site visitors. This can be as simple as finding more information about the company and its products.
Buyer Stage: Purchase
3. Convert – The final step that leads to the purchase either through online or at the offline channels. E-Commerce proceeds, product, price, and promotion.
Buyer Stage: Advocacy
4. Engage – Thrilled consumers are key to social media marketing. Long-term relationships to build consumer loyalty. From this step in the best situation leads back to the purchase part.
RACE consists of steps designed to help companies to engage with their consumers. It goes through the consumer journey that needs to be considered while planning digital marketing (Eriksen-Coats, 2019; Smart Insights, 2015). In this thesis, the engagement stage is the most important.
2.2 Social Media Marketing
Social media marketing (SMM) is a form of internet marketing that involves creating and sharing content on social media networks and allowing consumers to interact with the brand and with each other. SMM includes posting image updates, videos, and other updates like paid social media advertising (WordStream, 2019). Some of the most known social media platforms are Facebook, YouTube, and Instagram (Kallas, 2018).
For SMM to be effective, there should be a marketing plan. A good way to start would be by creating some goals that are wanted to be achieved. With SMM and proper planning, it is possible to increase brand awareness and build a strong brand personality, if the content that has been created is meaningful. It is important to monitor what kind of feedback the post-receive and respond to it. Sometimes fast reaction and quick-acting are needed because, in social media, things happen rapidly. SMM can also be used to alert consumers about what is going on and used to create a loyal fan base. The information of what are the best social media networks to use is critical. This way, the consumers or users are best reached, and reliable connections are made (York, 2018a). Concentrating on SMM is important; however, following competitors or some other companies that are doing well in their social media platforms is critical. What are they doing, and what kind of strategies are they implementing? What are the platforms they are using? There might be something that can be learned from them (Gurd, 2018).
In social media, many companies have created brand fan pages on social networking sites. On these pages, they can post videos, messages, quizzes, information, and other material. In these sites, the potential consumers and the already users can become fans of these pages and like or comment on them. This leads to commitment and loyalty, and these fans are normally more open to receiving information about the brand and more attached emotionally. Besides, liking and commenting reflect on brand popularity. Because of this, companies should experiment with different brands’ posts (De Vries et al., 2012).
2.3 Content Marketing in Social Media
Content marketing is something that is currently a hot topic of many articles and books. It has emerged mainly because of the change in the internet, social media, and technology that has made these easy to access and portable. The whole idea about content marketing starts from a strong knowledge of the consumers and then giving and bringing them extra value. With this marketing tactic, interesting marketing and content should be created, and this should also create brand awareness and gets consumers connected and engaged. According to Ahmad et al. (2015), social media marketing cannot function well without content as the content itself is something that the companies can share and post to the consumers. Good and valuable content can help to drive engagement in social media.
It is easy to get lost in social media because of the amount of information put there daily. Therefore, many things should be considered before marketing on social media. Some basic things would be the time of the post, type of the language, should emojis be used, and what kind of posts are made. Each of the social media platforms has its strengths, and too many of these facts are considered, there is not just one right answer. Many times, also the use of a few different platforms is the best solution, and this way, more consumers can be reached. Besides, this makes it possible to do more sharing, retweeting, and guiding consumers to access more information from different platforms. There- fore, knowing the consumers is the most crucial, and in content marketing, the quality is more important than the quantity (Verve Search, 2019).
Images shared on social media are seen as high value and have depth and density that words often lack. They represent communication that words may not be able to give. They allow more communication than most status updates or short sentences (Fris, 2014). The created content must be consistent and not to be afraid to leave the company to the background to realize some success. Brands should listen to their consumers, what they think of the current products or service, and what they would want in the future. This way, it is possible to create more value for the future post.
2.4 Effects on Consumer Behavior
It calls to understand the consumers and what is behind the purchase decision to create useful content. One of them is the Vakratsas and Amber model (1999) that goes through the different factors that influence the decisions (See fig. 3.).
Figure 3. A framework for studying how advertising works (Vakratsas and Ambler, 1999)
Everything starts with the advertising input and its execution. The second part is the filters that are the recipient’s motivation, attitude, interest, and the possibility to process the advertisements. The third one is the consumer’s emotional factors, previous experiences, gained information, and all this influence how the advertisement is taken. The last part is the behavior that comes from all things, such as education, religion, traditions, and habit-based purchase decisions (Vakratsas and Ambler, 1999).
According to Vakratsas and Ambler (1999), advertisements should be concerned with how advertising affects consumers, how it works, to formulate more effective advertising strategies. They claim that the emotional and previous experience is not valued enough and that advertising must have some mental effect, for instance, awareness, memory, attitude toward the brand before it can affect behavior. According to them, most products, and especially the frequently purchased packaged goods, are in the consumer´s mind already containing conscious and unconscious memories of earlier product purchases and usage, and this way, the attitude is not neutral (Vakratsas and Ambler, 1999).
2.5 Engagement in Social Media
People often turn to the internet for encouragement, entertainment, knowledge, and advice. The advice is based on what to buy, where to buy, and who to buy from (Leary, 2008). Social media is a great place to promote a brand, but still, it has the challenge of being a platform where brands need to be continuously active. Consumers want information, service, help with their problems, and the feeling of connection. Consumer engagement is about encouraging the consumers to be connected and, as a brand, offer some support, guide, and even honor them. When this is done well, it will help the company and the brand to grow and create more loyal consumers.
The key is to bring value and create engagement. Consumers want a fantastic experience, and it is just not while they are buying the product or service, but it should be a long-lasting bond. With consumers being able to relate to the brand and by creating a relationship, the brand can gain valuable information about their consumers. By finding the needs and wants, it is possible to understand any changes and improvements that should be made. Brands should be interested and try to get interaction with the already users and potential consumers. Sometimes the response or comments does not hold or might be hard to take, but the critic should be taken positively. The end-to-end experience is the most important, and many times the brand can do something, even if something has gone wrong or is bothering the consumer. Besides receiving information, the brand is also able to educate the consumers. By finding the right platforms and the right kind of posts, the brand can tell the consumers what values they represent and create valuable content. By getting the consumer’s trust, loyalty is gained too (Ference, 2017).
In measuring the return on investment from engaging consumers across the social media platforms, Lipsman et al. (2012) point out the value of online users and fans in improving the incomes. The authors note that the degree and extent of consumer engagement across the online platforms form the basis of measuring the potential return on investments. In many cases, reaching out to the potential customers provides a basis for determining their levels of satisfaction for a product. Thus, learning how online consumers devote their time during the engagement period provides a basis for establishing the non-monetary returns on investment. Additionally, Paine (2011) expresses the importance of understanding the increasing purchasing behavior among consumers. The actual sales often result in increased returns on investment while ensuring that the satisfied customers can attest to their satisfaction from the products. The ability of social media users to influence their friends forms the basis of evaluating the success in higher returns on investments (Vlachvei & Notta, 2015). Besides, Pradiptarini (2011) emphasize the critical elements for measuring the success of return on investments. The elements include improving the experience of the social media content, building lasting customer relationships through involvements, and integrating the usage of diverse social media platforms. Thus, measuring the return on investment should focus on particular attributes of consumer engagements across the interactive media platforms.
Figure 4. Facebook Engagement Pyramid (Sharma et al., 2017)
Figure 4. represents the levels of Facebook engagement. This same pyramid can also be used for other social media platforms. From the bottom, there is liking, which is easy for many consumers but involves little engagement. The next level is the reaction, which represents the emoji that mostly stands for a feeling that the post has caused. This takes more effort than liking, but still, it is easy and does not consume much time. The third stage is the comment that needs more effort and engagement. The fourth stage is sharing, which is a high-level engagement that shows trust in the brand. Create is the last stage where consumers start creating their content for the brand (Sharma et al., 2017).
Another way to analyze the levels of engagement would be the COBRAs concept that stands for consumers´ online brand-related activities. It is a concept that provides a frame for consumer activity towards brand-related content on social media platforms. It studies and compares online behaviors. COBRA categorizes the users into three usage types, i.e., consumer, contributor, and creator. The consumer uses the minimum level of activeness, which includes reading, watching, and following what the brand or the other users have created. The contributor is the middle level of activeness through comments, product rating, engaging in conversations, and joining groups. Finally, the creator type who are ultimate on brand-related activeness. Creators write brand-related blogs, creates videos, and writes product reviews (Muntinga et al., 2011).
Blanchard (2011) suggests that combining social media analytics improves the ability of a firm to track its performance following the anticipated returns of its investments. The study proposes that the calculation of the ROI on social media marketing should involve a simple approach in assessing the company’s sales resulting from the promotional links and advertisement media enhanced through its websites and social media platforms. Through these aspects, firms can establish how their social media marketing impacts on their investments and returns. Powell et al. (2011) note that through the use of social media frameworks, it is possible to measure the impact of the growing marketing trends across the online platforms.
In the framework, understanding consumer behavior and response provide an opportunity to measure the investment returns in line with the social media responses. Rust et al. (2004) point out the importance of determining consumer awareness, attitudes, experience, and associations in measuring the returns on investment. Thus, firms need to consider various aspects of consumer purchase and experience in determining the income they would receive from their investments. Through the idea of marketing productivity, Rust et al., (2004) highlight the need to foster consumer experience by increasing brand awareness and translating the knowledge in sales for their profitability.
Additionally, Sterne (2010) posits that social media is an opportunity through which firms can realize growth and increase their productivity across the global marketplace. The author highlights the need for organizations to adopt a real-time strategy of determining the impact of their product by querying the consumers regarding their experiences with a particular product. As such, firms can increase their returns on investment by gearing to evaluate their business prospects and values through strengthening their relationships with their consumers on online platforms. Tiago and Veríssimo (2010) express the growing need for firms to adopt sustainable means of advertisement that would improve their economic incomes. With digital marketing complementing the traditional means of advertising, it is increasingly becoming necessary for market survival.
Again, Tiago and Veríssimo (2010) note the essential metrics for measuring the returns on investment include; information gathering processes and feedback from the social media platforms and websites. Surveying participants to assess their potential purchasing power goes a long in determining the prospective returns from an investment. Alves et al. (2016), through the analysis of various studies, point out the need to evaluate the benefits created by social media marketing towards improving the returns on investment. Throughout the studies, it is evident that determining the customer responsiveness by positive or negative feedback provides a rationale for improving the investment prospects. Perdue (2010) points out that a firm can attempt to enhance its competitive advantage as a way of increasing the profitability of their market investments.
Return on engagement (ROE) is the new form of return on investment (ROI). This results from a rise in customer-based business operations in organizations. The changing tastes and preferences of customers have to be managed for a given brand to have effective coverage in the market. Return on engagement is an essential social media engagement metric since it serves a purpose that is greater than what the return on investment has been doing. Whereas, return on investment focuses on numbers, return on engagement focuses on successful customer engagement. It is the strength that a brand gains from different brand actions, strategies, or products.
Despite being difficult to measure, it is more valuable than the numerical values on which the return on investment is based. Cristea and Mocuta (2018) argue that social media platforms, such as Facebook, have millions of people interconnected. There is a virtual community on social media, and it is in this community that marketing activities are conducted. They appreciate the essential role that social media is playing in the marketing of goods and services over the internet. They continue by analyzing the different measures that can be taken by a brand manager to ensure effective customer management is achieved through various social media platforms.
Building and investing in relationships is an essential factor in the development of competitive advantage in business. Social media is a base on which such relationships can be built because of the nature of engagements that different people have. According to Gopinath, Thomas, and Krishnamurthi (2014), social media is the foundation of business growth over the past decade. They argue that the content that is available online is essential in building or destroying brands. Managers need to have highly effective strategies in place for their brands to be effective in the market. Gopinath, Thomas, and Krishnamurthi (2014) argue that online word of mouth has been essential in increasing sales levels of different organizations. Word of mouth used in online platforms depends on the kind of relationships that are developed between the companies and their customers. If the company creates a bad reputation or some of the customers have bad experiences with the company, they are likely to spread negative information about such brands that damage the company’s reputation. Whenever the brand’s reputation is affected, such a product is likely to fail in the market.
Killian and McManus (2015) conducted a study on marketing communication, and they were comparing traditional forms of marketing and online-based business operations. They conducted an evidence-based survey on how the consumer market has shifted and found out that the highly computerized organizations and those that engage in online campaigns had an advantage over those engaging in traditional marketing communication. They continue stating that social media platforms offer an easy channel through which marketing communication can be conducted. Killian and McManus (2015) posit that managers tend to categorize social media platforms into four, where there is relationship management, creativity, newsgathering, and entertainment. Reactions from different customers on these platforms are essential in the satisfactory delivery of products and services to the market.
Brand awareness is developed through social media platforms where the users of different brands can show their reactions and give feedback on their experiences with different brands. Zhu and Chen (2015) have similar arguments that social media has become an essential marketing platform that should be considered. There is more engagement with customers since there are always people on social media platforms. They argue that the technological advancements that have come into play are significant in business, and the world has been globalized through on-line business engagements. Hajli (2014) argues that social media has an impact on consumer behaviors. Managers need to monitor the activities and trends that are popular on the different social media platforms for them to understand the market needs. Some trends arise through engagements in social media. Buyers and sellers are connected on social media easily, and that is why there is a need for such sellers to be careful not to hurt their reputation in this large community. People on social media tend to influence each other, and it is through such influence that different groups of people can come up with different tastes and preferences. The trends that are adopted by customers can affect the way they relate to a given brand, and that can promote or destroy a brand depending on how they relate with such a brand.
Xu et al. (2017) argue that social media utilizes the internet, and different people in American society can communicate effectively. It does not matter whether they are in the urban or rural regions of the country. The established social media communication network has led to the development of effective markets that have promoted different businesses across the globe. Shopping has become more of a fun activity since people can view products analyze them as well as make decisions on whether to make purchases or not (Scarpi, Pizzi, and Visentin, 2014). There is a lot of information on social media, and it is playing a crucial role in understanding the market and the products that people need to consume.
Consumer behavior is a factor that should be considered in the analysis of online shopping patterns. Lim et al. (2016) analyzed the factors that influence online shopping behavior among different people. Intent and social media play a vital role in shaping the behavior of online shoppers. The intention to make a purchase should be clear and the person engaging in the shopping activity should have a clear mind of what he or she would like to buy.
There are other factors like social media that influence the decisions made by the purchaser. Social media has trends that are created by different people who can have a wave of influence on a given product. Such trends affect individuals because they influence their decisions on the purchases they would like to make. Kostyra, Reiner, Natter, and Klapper (2016) argue that online business has led to high levels of competition. It may hurt businesses, but it enables customers to have a wide variety to choose their products. Consumers have positive perceptions and attitudes towards online shopping because they believe the platforms offer high levels of convenience that are needed in their busy schedules (Jiang, Yang, & Jun, 2013).
2.6 E-Commerce
E-commerce has been sustained by social media platforms. Most of the e-commerce business entities tend to relay a lot of information about their products on social media. Zhu and Chen (2015) posit that the businesses have been improving over time because of their presence on social media platforms, and it has been an essential factor in their development in different markets.
The websites have links to different social media platforms, and it has been a primary factor in improving their engagements with customers as well as marketing their products. Tiago and Veríssimo (2014) discuss the development of e-commerce in the US over the past ten years and how social media has played an essential role in ensuring that marketing is effective. Organizations today have social media managers with the sole responsibility of creating relevant content that can be used to improve their brands. Managers are aware that the virtual community that has been created by social media is conscious. Whenever there is a negative feedback from such people, there is a need to work out ways to ensure they are satisfied (Killian & McManus, 2015).
Businesses have benefited from online engagements because of the wide market coverage that has been achieved through various social media platforms. Shanthi & Kannaiah (2015) argue that globalization has been achieved with increased internet use. Online shopping is increasingly becoming popular, and different businesses have the advantage of using the internet to access customers from different parts of the world. According to Shanthi and Kannaiah (2015), companies like e-bay have gained a lot of popularity and business opportunities as a result of engaging the virtual community that is available in the online social media platforms. They posit that internet shopping has its advantages that have helped different companies to develop in the market. Online stores are increasingly becoming popular as customer perceptions change positively, and confidence develops. The internet links people from different parts of the world that allows businesses to access the market through the virtual world. When a brand gains access to new markets through social media, it is one of the benefits that the business gets, and it is the return on engagement. When customers in new markets are engaged effectively, they tend to respond positively towards brands.
The internet is another essential factor that has led to the effective development of different forms of business interactions. In the twenty-first century, the corporate world has come to realize the importance of the internet that has led to a paradigm shift in how business is conducted. The internet enables systems to exchange data without physical interaction between people. For instance, an individual can make an order, and the automated system will carry out the transaction to completion. Such engagements have improved customer services, and they have been essential in promoting brands across different regions. People are increasingly accepting the use of the internet in carrying out different business transactions since they offer smooth services. Social media has played an important role in the marketing function of the corporates since they can utilize the platforms to popularize their brands as well as engage their customers and potential customers. Any business that does not use the internet today is at a disadvantage because some people do not have time to analyze the information available on traditional media platforms (Dijkman et al., 2015). They use the internet all the time and visit social media platforms regularly, and therefore, marketing is effective through the platforms.
According to Hortaçsu and Syverson (2015), there is an ongoing evolution of the retail industry in the USA. The authors analyze different forms of data that relate to the subject of e-commerce in the country. They note that Americans use the internet daily in various activities of their lives. The authors state that at least 70% of the population has access to the internet, which has led to the rapid development of e-commerce in the country. Besides, there are electronic devices that have been developed, and they are user-friendly. Smartphones, laptops, and computers have been made available that makes the use of the internet easy for all in society. Most Americans find it convenient when they use online platforms in making their transactions. The accessibility to affordable electronic gadgets and the internet has made it possible for the citizens to embrace e-commerce.
Different industries ranging from electronics to groceries have developed online platforms from which they can make deliveries to their customers. Krotov (2017) supports these arguments by stating that the Internet of Things (IoT) has come in handy in the development of the e-commerce business. A mutual relationship has been created between entrepreneurs and their clients through these technologies. Customized service delivery is achievable due to the available technologies that make it easy for businesses to interact with their customers or potential clients. Social media platforms have enhanced the popularity of e-commerce businesses since most of their websites have links to different social media platforms.
The internet has led to the diversification of the business sector. Lee and Lee (2015) argue that the internet has played a significant role in helping businesses to access new markets that they would not be able to access without the use of the Internet of Things. Business entities are increasingly transacting millions of dollars through online platforms that have proven to be highly effective and accepted among many people in the world. Reviews that are given on social media platforms play a vital role in making the brands popular, and it is through such engagements that the businesses have become more productive. Without the internet, the companies would not have diversified their markets because it would not be easy for them to carry out transactions in such markets without online platforms. Miller et al. (2018) posit that the Internet of Things and e-commerce is not only beneficial to sellers but buyers too. Buyers have a great source of products that they would like to purchase. The e-commerce platforms and social media have allowed the buyers to analyze different brands and settle on the ones that they are comfortable with (Hortaçsu & Syverson, 2015). The assortment of brands on the internet has led to high levels of competition that has made companies work extra hard to satisfy their customers’ needs.
According to Mosteller, Donthu, and Eroglu (2014), online shopping has various benefits to shoppers. The researchers conducted an evidence-based study on how different people view online shopping. People have developed a habit of using different online platforms, and they find them more efficient since they need to engage in other activities and save time that could have been used for shopping. The platforms are highly valuable because they utilize the available information about their customers, and that makes it easy for them to deliver the services effectively (Huang & Benyoucef, 2015). Sellers can analyze the market needs and develop the most effective strategies that can be used to provide such services to satisfactory levels. The smooth online shopping experience is what the shoppers would like to have in their shopping needs. When the shopping experience is less hectic and involving, it makes it possible for the different people to handle other duties more effectively.
When the online shopping platforms were beginning, most people did not trust them. It was a revolution that was coming to the retail industry, and people were not ready to accept it (O’Reilly, 2012). However, today many people prefer to shop online because the platforms have proven to be safe. It has encouraged more companies to take this route in business. Social media has played a useful role in the marketing of brands where they easily share information about their products, and potential customers can view and give recommendations.
The US Department of Commerce (2016) reported on how e-commerce has picked in the US, and most of the US citizens have shifted from the traditional forms of business engagements, and now they embrace e-commerce. According to Wylie (2014), the social media revolution has played a critical role in the development of e-commerce. The more people engage on social media, the more they are attracted to online transactions that make it an essential tool meant to promote e-commerce as a highly profitable business model. The return on engagement has been evident in the use of internet platforms. The efforts have been effective in delivering a highly effective business model through e-commerce platforms. Zhou, Zhang, and Zimmermann (2013) argue that the use of e-commerce is indispensable since people have come to embrace it as a means of accessing different products efficiently. The society wants efficiency and effectiveness, and e-commerce has been able to deliver services in the way the customers prefer.
2.7 Instagram
Instagram is full of pictures of dreams, life, and products. Nowadays, Instagram has expanded to the use of clips, stories, videos, and live-programs; however, the most prominent thing is the pictures. Based on Instagram’s research, it is a platform where people express themselves visually and are connected to the things that are important to them (Business Instagram, 2017). Instagram, at the moment, has around 1 billion monthly active users and more than 25 million business accounts (Sharma R, 2018). Also, 60% of users visit the platform daily. From all the users of the internet, 32% are on Instagram (Sharma G, 2018).
Instagram is a significant platform for users to share media, reflecting their interests. It is used by marketers and brands to reach their potential audience for the advertisement. The number of likes on posts serves as a proxy for the social reputation of the users. In some cases, social media influencers with an extensive reach are compensated by marketers to promote products. This emerging market has led to users artificially bolstering the likes they get to project an inflated social worth. In this study, the researcher enumerates the potential factors which contribute towards a genuine like on Instagram.
2.7.1 Instagram History
Kevin Systrom and Mike Krieger created Instagram as a smartphone application. When it was launched on 10th June 2010, 25,000 users showed up the first day and one million users in three months. With its rapid growth, the investors got interested in the platform, and Twitter tried to purchase the platform. Later Facebook also made an offer, and Instagram was purchased for 1 billion dollars in 2012. Instagram has been developing new features and versions, for instance, in 2013, a feature of tagging of photos and introduced videos. The key to Instagram’s success has been claimed to be the easiness of its use and that people do not want to read a long text but prefer watching videos and pictures (Izood, 2018).
2.7.2 Instagram Use
Instagram is easy to download from the App Store or Google Play Store, depending on what kind of digital device is used. Once it is installed, the user must sign up or create a new account to use the platform. It can also be used from a computer, and therefore, not necessarily to be used in smartphones as initially developed. A personal profile is needed to share photos, like, and comment on Instagram. The profile works based on a username. Instagram usernames are provided on a first-come, first-served basis, and usernames may not be reserved. The same rules apply to private persons and companies (Instagram 2019a; Instagram 2019b).
Instagram offers the possibility nowadays for companies or even private persons to switch their profiles into Business Profiles. With business profiles, companies can access new business features and Instagram Insights. Instagram Insights gives information on who are the followers, when they are online, among others. Also, more details of the post are available and data about how they have performed. With a business profile, companies can add contact information and a category where the company belongs (Instagram, 2019c). Another valuable feature on the business profile is that it helps the user to create ads like the feature on Facebook, or promote the already existing stories (Instagram, 2019d; Instagram, 2019e).
There are many options to post an item on Instagram. There is the possibility of posting a picture or many pictures simultaneously; however, it has a limit of 10 pictures. There is also the choice of a 60-second video or the same number of videos in one single post. The pictures and the videos can be mixed in the same post. Besides, short video clips can be added with a maximum of 15 seconds that become the user’s story, which will vanish after a day unless they are saved. Another option is to do a live video show, or use the newest addition IGTV (Instagram Television) where the videos can be between 15 seconds to even 10 minutes long. Instagram offers the possibility to edit the pictures or videos and offers many of its filters (Instagram 2019f; Instagram, 2019g).
2.7.3 Brands on Instagram
On Instagram, 80% of the accounts follow some company to create a personal connection with the company. Also, according to Instagram’s studies, the consumers’ loyalty is based on what sort of feeling the brand creates, and many people use and follow the popular hashtags. They claim that many people turn in to Instagram to search for inspiration and search products and services. It is an easy way to feel like part of a group, stay connected, and have a feeling of making an effect (Business Instagram, 2017).
2.7.4 Brand Exposure on Instagram
Instagram has the third most users after Facebook and YouTube. It is an easy way to be connected with consumers, even above a country’s boundaries. It is a global platform for products and services (Clarke, 2018). In most cases, Instagram is mostly used between the ages of 18-34. However, as the use of social media is growing, many older people are joining the platform (Kemp, 2018). Between the young users from 13 to 17-year-old, it is the second used platform after Snapchat (Clarke, 2018).
On Instagram, 60% of its users have said that they find new products from Instagram (Business Instagram, 2019a), and 72% of users have bought a product they saw on Instagram (Keys, 2017). In addition, over 2 million Instagram users visit at least one business profile in a day (Business Instagram, 2019b). Instagram is a platform where the consumers and potential consumers can get details about the brand and what it represents. For companies, it is possible to get real-time data and information about their followers. It is an easy way to stay in contact with the consumers and get valuable information from them and, at the same time, to inform and provide extra information to consumers. Instagram is easy to access and easy to use (Business Instagram, 2019b).
2.7.5 Brand Posts
As mentioned under Instagram Use above, the platform offers the possibility to post different types of items. There is a picture, short video clip, story, and IGTV posts. While doing various posts, the knowledge of the followers becomes essential, for instance, what kind of post they might like? Also, it might be a good idea to search for what the competitors are doing, the type of post they are using, and the content. Every post should have a goal, and while doing a post, the focus should be the followers (Daley, 2018).
Instagram is very visual, so while sharing posts, they should be styled in a way that represents the brand and is visually interesting. A good post might be a product use post, that shows good ways to use the products. Also, a post can be a great way to update followers on products, i.e., is there something happening, some new products coming, or perhaps it can be about some special promotions or sales. Another good post might be just an inspirational post. Not all have to be about products, service, or purpose, but they can be just inspirational and create feelings. A post can be a suitable quote or a celebration post. Some milestones and holidays are an excellent time to post something different and even perhaps with some humor in it (Pilon, 2018).
When the brand engages with the followers, the brand might find some interesting posts that the follower has made about the brand. Such posts can be reposted as a sign of appreciation and to show the followers that the brand cares and has an interest in their contents. This content is called user-generated content (UGC). According to L2 Intelligence Report about the Instagram, consumer who views UGC is 4.6% more likely to buy the product, and a consumer who interacts with UGC is 9.6% more likely to purchase the product (L2, 2015). This indicates that reposting brings value to a brand.
Another way to get the followers excited is some sort of contest or giveaways posts. The contest is an excellent way to get new followers by asking them to post something or hashtag the posted photos and videos. As such, the followers can take part in the contest, and at the same time, creating more content under the used hashtag. The followers of the followers can notice such posts, and therefore, adding more value. Another option for a post is to ask the followers opinions of products, service, or its use. The followers should be challenged to do something, and they should be activated and motivated. A good post might also be telling the followers more about the brand and not just about what the brand is offering. Brands should post what is going on behind the scene, brand changes, and perhaps make some posts about the employee. As such, they will make the followers feel part of a big family and know what happens behind it all (Pilon, 2018).
Figure 5. Brand Actions on Social That Prompt Consumers to Purchase (Chaffey, 2019)
In figure 5, from Sprout Social, brand actions that prompt the consumer to the purchase decision on social media can be seen. For brands, this gives a good guideline on what to do and on what the posts could contain. The top things were being responsive, offering promotions, providing educational content, and sharing exciting visuals (Chaffey, 2019).
2.7.6 Photos and Videos
According to Mawhinney (2019), visual content is becoming more critical. Pictures are worth thousands of words. People following instructions with an illustration do much better. If information is heard, only 10% of that information is likely to be remembered three days later. However, if an image is paired with the information, 65% can be remembered after the same time. The picture is more powerful and faster to comprehend than words (Medina, 2019).
Videos have become a powerful tool, and it is an easy way to communicate. Virtual Reality (VR) is becoming more useful and establishing its place in the platform (Mawhinney, 2019). Stelzner (2018) indicates that 32% of marketers say visual images are the most important form of content. 80% of marketers use visual assets in their social media marketing, while 63% use videos. In a 2018 HubSpot survey, 54% of consumers wanted to see more video content from a brand or business they support (Mawhinney, 2019).
The popularity of videos has grown dramatically and keeps on growing. According to Wyzowl´s “5th annual State of Video Marketing Survey in 2018”, 81% of businesses were using videos as marketing tools, and 87% of consumers say that they would like to see more videos from brands (Hayes, 2019).
2.7.7 Likes and Comments
A big part of using Instagram is liking the posts and commenting on them. Brands should not just want to increase the number of followers but want the followers to be active through the use of likes and comments. This creates the engagement as it was already earlier accomplished. Around 3.5 billion likes are given daily on Instagram (Pathak, 2018). On Instagram, comments and likes on the posts represent the popularity and good engagement. If there are not many likes and comments, despite a large number of followers, the brand’s strategy should be re-evaluated. In that case, the quality of the content and ways to engage the followers should be changed. Brands should try to listen to the comments and see the posts get the highest number of likes. Comments on the posts should show that the audience is responsive, trusts the brand and that the content is relevant. A lot of engagement shows that the followers are interested in the things that the brand offers (Tucker, 2018).
Consumers who are liking an image of a brand are inviting the brand to their lives and following the brand´s creativity, which creates more of an emotional connection. There is not a specific format for a picture that would get more likes, but based on a study made by Curalate on 8,000,000 Instagram images, there are few things that stand out. In a picture, high lightness generated 24% likes more than dark images. Pictures with a spacious background attracted 29% more likes than those without such background. In addition, images with blue as the dominant color generated 24% more likes than red images. Images with a high level of textures generated 79% more likes than those without the texture (Lowry, 2013). The higher the number of audiences engaged, the higher the chance for them to become followers and, eventually, brand consumers. With their likes and comments, the brand can collect valuable data about products and posts. There is much to learn from the audience about brands (Daley, 2018).
2.7.8 Post Time and Frequency
Understanding what time to post is vital for companies because the posts target as many users as possible. As such, it is worth noting how Instagram works to understand the time and frequency of posting items. All social media platforms normally have some kind of algorithms that defines how they function. Instagram has not been that open about telling how they run things, but during the past few years, they have changed how users can see the pictures and the algorithm behind the platform (Constine, 2018).
Instagram claimed that earlier when they had a chronological feed, users were missing 70% of all the posts. However, they changed in 2016 to a feed that shows the posts by the user´s past behavior and evaluates the user’s interest. It is based on how users interact with the accounts, tagging and commenting, and based on similar content. Also, the recency effects have been added so that it prioritizes the newer posts, but if the user keeps on scrolling, all the posts will end up showing. The feed does not favor pictures instead of videos; however, it is based on the user’s past behavior. It does not matter if a special feature is used, is it live or story, or is the account business profile or personal. If several posts are made from the same profile one after another, it will separate them (Constine, 2018; Leppänen, 2018).
Instagram users engage more on weekdays, especially on Tuesday and Thursday (Sharma G, 2018). The algorithm supports the activity of businesses and their followers. Brands should post the kind of pictures and videos that many users want to see and interesting ones. By being active, getting the followers to comment and like, taking part in the conversations, and liking back, the brands increase their possibilities of their posts appearing in the top list son several occasions. Based on a study of Tailwind that studied over 100,000 Instagram posts from three months, it showed that there was a significant difference in engagement rates between those who post once per week between those who post every day (Christopher, 2017). This indicates that companies should post frequently.
2.7.9 # Hashtags
Hashtags were created in 2007 by Chris Messina, a former Google developer, but they became well known after Instagram introduced them in 2011 (Pathak, 2018). They are sort of keywords marked with a # and are nowadays used broadly. On Instagram, their purpose is to link a post together with all the posts that have the same hashtags. As such, posts and items can be searched using the hashtags.
Using hashtags is also useful for the brands because a post with at least one hashtag creates an average of 12.6% more engagement (Sharma G, 2018). There are two types of hashtags, short-tail hashtags, and long-tail hashtags (See fig.6). The short tail cannot be more than two words long, while the long tail is all the hashtags with at least three words. The short tail has a lot of competition, as many of them are often used. As such, there is a chance that the post with short tail hashtag gets lost among the other posts. The good thing with the long tail is that it is more specific and leads to more exact search results. It is more targeted, and typically there are not so many search results (Chatz, 2017).
Branded hashtags are specific to a company. They can contain the brand, products, or a particular campaign (York, 2018b). Seven out of ten hashtags on Instagram are branded (Sharma G, 2018). Then there are the general community hashtags that can be anything and are not tied to a specific thing. When doing a hashtag, it should always be relevant to the post and easy to find it using common keywords. There should be some variations in them because this way, a wider audience can be reached. It would also be good for the brands to check what shows up when these hashtags are used (York, 2018b).
Figure 6. Short tail and long tail hashtags (Pathak, 2018)
2.7.10 Mentions, Tagging, and Influencers
When doing an Instagram post, it is possible to mention someone in the text by adding @ before their profile name or tagging the profile to the picture. This means that if the picture is clicked, it will show the profiles tagged. For example, if there is a picture of a room, the furniture’s can be tagged based on the place of purchase, or mentioned on the text by adding @ before their profile name. This way, all those who view the pictures can go straight to the profile pages of the mentioned or tagged brands just by clicking on them (Instagram, 2019h). Instagram also offers the opportunity for companies to tag their products on their posts, and through a click, it leads to the company’s website (Business Instagram 2017).
For the companies, mentions and tagging can be a good way to get attention and spread their products or services. It is advisable to encourage the followers to tag the brand to their photos and mention them in their comments, along with using the hashtags. Good ways to encourage them to do this can be campaigns or contests (Parker, 2018). Another option to spread the word is to use some social media influencers. Sometimes the brand might get lucky and be mentioned without any effort from the company. The company can also try to organize some marketing campaigns with the influencers and do a collaboration. The best would be to get the influencer’s genuine interest and to get them to tag or mention the brand more than once because not everyone will notice the first post (Verve Search 2019). The brands should try to find the influencers that have the followers that could be suitable based on the target market. In 2017, 92% of marketers who used influencers found it effective (Forsey, 2019).
Influencers are users who have established credibility, large numbers of followers and can affect others to act based on their recommendations. They can attract many viewers consistently and can motivate their followers. An influencer can have multiple backgrounds as a blogger, celebrity, or an online entrepreneur. The audience is not limited to their actual followers, but they can connect with the followers of their followers by getting the followers to share the content. Many social media influencers get paid for what they share on Instagram. The payment is because they can reach a lot of people and get more social engagement (Rivera, 2019).
2.7.11 Engagement on Instagram
The algorithm on Instagram supports engagement. By staying connected, the brands have a better possibility of reaching the audience and potential consumers. It is just not about getting the likes and comments, but also responding to them. All the aspects of a post can help the brand to succeed, so the hashtags and the emojis should always be included. If the social media strategy is successful, this can be noticed in the engagement rates. Content can bring engagement if it has valuable information, but it can also do it by being just enjoyable and by bringing joy (Calder et al., 2009).
Instagram does not have its engagement measurement but can be analyzed from the interest and interaction shown towards the posts. Even if there are a high number of followers, the engagement might be low. Just tracking the number of brand’s followers is not enough. However, it is important to determine how frequently the new followers are gained because this shows that the content is interesting. The number of followers should be compared to the number of likes and comments a post is getting (Mintells, 2019). A relatively high number of likes and comments in relation to the number of followers is an indication of an excellent piece of work. However, if the responses are low, the content creators should work on improving the posts.
In figure 7., it can be seen how Instagram engagement can be calculated. It should be counted to each of the posts separately. This formula will be used in the empirical part to calculate the engagement.
Figure 7. Instagram Engagement Rate (Mintells, 2019)
In figure 8., “The Average Engagement Rate of 2018” on Instagram can be seen. The rates depend on the number of followers, and the rates change over time (Komok, 2018).
Figure 8. Average Engagement Rate on Instagram (Komok, 2018)
2.7.12 Fake Likes on Instagram
Apart from being used as a medium of communication, Online Social Networks (OSNs) are also used to gain popularity, increase social self-worth, and promote businesses. Even brands, advertisers, and the background recommender algorithms of OSNs rely on the popularity metrics of users and content shared on these services. To get more benefits, users often artificially increase the popularity and engagement of their content in several ways. Some of the conventional methods are to leverage bots, purchase social metrics such as likes, followers, and shares from black market services and become part of collusion networks, which can be used to trade in organic engagement. Such artificial bolstering of popularity can result in loss of money [1] by brands because advertisers to not reach the relevant audience, and recommend algorithms to give poor suggestions [3].
There have been several prior efforts to identify fraud [6], spam [4], and fake users [6] on OSNs. In this study, the researcher instead focuses on inorganic engagement received by a user. Previous studies aiming to detect fake liking behavior, assume that if a user has given one or two fake likes, all her likes are fake [2, 8]. However, we believe this is a limited understanding of fake engagement since a single user can generate organic, as well as inorganic engagement. For instance, an Instagram user can like content with some genuine interest. Besides, the same user can also be a part of a colluding ‘like-back’ network, where she likes unrelated and random content only to receive back some likes and increase her popularity. Therefore, the researcher proposes that the true reach or social-worth of the user should be determined by canceling out the effect of fake engagement, which the user receives, and should largely depend only on the organic engagement (the terms organic and genuine will be used interchangeably). The researcher defines organic liking engagement on Instagram as a like-action which a user gives to a post with some genuine interest in the content, or in the user posting the content (poster). In this study, the goal is to identify the ingenuity of likes by determining the user’s intention of liking a post. In particular, the researcher defines the goal as a user who likes a specific post p of a poster, finds out the features of, p and, determines the probability of the user genuinely liking the post p.
Here, the researcher finds attributes of fake liking, which can help distinguish such behavior from organic liking activity. Unlike previous studies on spam detection which assume that a single spam post translates into the user being a spammer [2, 8], the study infers the reach of an Instagram user as a function of the organic and fake likes, reducing the effect of fake likes (to some extent subject to the efficacy of the approach) in the process. The contributions are Characterizing Fake and Organic Likes. The fake and organic likes are discerned by determining the factors which can lead to a user genuinely liking a post on Instagram. While previous studies have looked at meta-features of profiles, content, and structure, the study focus on identifying the probability of a user liking particular content based on various factors like topical interest and proximity with the source user. The research will involve studying an extensive list of features indicating genuineness of a like. The research findings show that the topical interest of a liker with the post and profile quality is helpful. The application of Automatic Detection of Fake Likes will be the next step; however, it will be based on the understanding gained in the previous step. The researcher will build a machine learning-based model to distinguish a fake like from an organic-like automatically. The researcher can detect fake liking instances with a precision of 83.5% using a neural network model.
Chapter 3: Methodology
3.1 Research Methods
Social media sites play an instrumental role in both development and advertising in business as they enable brands to reach their targets through advertisement, involve clients as well as to conduct a market exploration; however, the major drawback is internet access. The internet plays an integral part in forming an interlink between clients and the internet. There has been a complete exploration of the usage of social media from a business viewpoint. Some studies on return on investment have not fully taken off. ROE, in relation to social media, is a new concept and is being attributed as the perfect performance metric that could measure optimistic outcomes of customer communication through social media. Thus, it is authoritative to discover the rationality of ROE as a social media engagement performance metric, assess its influence, and, in the end, evaluate its results. There are eight major classifications of research designs. These include observational research, correlational research, true experiments, developmental study design, survey study design, validity, statistical conclusion validity, and internal validity. Each of these will be discussed further below.
3.1.1 Observational Research
As the name suggests, this design involves directly observing individuals in their natural habitat or setting, and that’s why it is occasionally referred to as a natural experiment (Rosenbaum, 2002). In other words, it describes the type of study where the exposure being evaluated is not assigned by the investigator, which suggests that no randomization occurs during the experiment. Regarding the latter, it is also referred to as a quasi-experiment since the selection of the participants must be conducted in a way that it enhances the validity of the research question (Rosenbaum, 2002). Observational studies have also been known to be prospective, cross-sectional, or retrospective, which is all determined by the exposure, and therefore if it was recorded before the outcome of the study occurred, then it can be deemed as being a prospective study. The retrospective study, however, results when the exposure is collected after the occurrence. A cross-sectional study, on the other hand, occurs when both the exposure as well as the outcome are determined at the same time (Rothman, Greenland & Lash, 2008). Other studies that can be classified under this design include ethnographic studies, ethological studies, case studies as well as some of the survey studies since they involved a collection of data through observation and recording.
Often observational studies are qualitative. The social media case study will entail extensive notes based on observations of and interviews with various social media stakeholders such as influences, customers, marketers, among others. A detailed report with analysis will be written and reported constituting the study of the scenario used in the research. These studies may also be qualitative or include qualitative components.
3.1.2 Correlation Study Design
In any given research, the investigator, apart from the study problem, is always interested in understanding how two or more variables are related or how they affect one another. For instance, if one needs to determine the variability of blood pressure (B.P) by race and exercise activities involved, the latter would be the independent variable while the B.P will be the dependent variable. Correlation study design, therefore, involves non-experimental research where a researcher measures two variables and, after that, assess them to determine the statistical relationship that exists between them with no or little effort to control the extraneous variables (Well & Myers, 2003). This design is similar to observational study design since they are both non-experimental and involves the collection of empirical data through observation of the subjects. The design also compares to the observational one since they both believe that the statistical relationship being investigated is not a causal one. However, unlike the correlation study design, observational studies do not involve analysis of variables to test for significance; for instance, it requires measurement of duration for each observed trait of the subject. Both, on the other hand, ensure that the research cannot manipulate the independent variables of the study itself because it is impossible, considered unethical and impractical (Shuttleworth, 2008). The measures for the case of observational research can be qualitative, while that of correlation can be quantitative when the collected data is analyzed. Furthermore, correlation can be termed as negative (-1), positive (+1), and nonexistent (0), which cannot be said to be the same case as in observational study (Devi Et al., 2018).
3.1.3 True Experiments
A true experiment is defined as an experiment conducted where an effort is made to impose control over all other variables except the one under study. To understand the nature of the experiment, the researcher must first define a few terms:
1. Experimental or treatment group – this is the group that comprises organic social media users, e.g., users that freely follow, like, and share posts and information on social media. The group receives the case studies that are different from the control group: Bought Social Media Services, e.g., both followers, fans & likes on the variable under study.
2. Control group: Bought Social Media Services – this group produces comparisons. The treatment of interest is deliberately withheld or manipulated to provide a baseline performance with which to compare the experimental or treatment group’s performance.
3. Independent variable – this is the variable that the researcher will use to manipulate the study. It can be any aspect of the environment that is empirically investigated, e.g., buying social media services, to examine its influence on the dependent variable.
4. Dependent variable – This variable is dependent on the independent variable. As the independent variable changes, this will also result in a change in the dependent variable. Therefore, the researcher will observe and record the results.
5. Random assignment – in a study, each subject has an equal probability of being selected for either the treatment or control group.
6. Double-blind – neither the subject nor the experimenter knows whether the subject is in the treatment of the control condition.
As stated previously, the control group will provide us with a baseline for comparison. All subjects will be randomly assigned to groups, be tested simultaneously as possible, and the experiment will be conducted through double-blind. Perhaps an example will help clarify these points. The researcher will create accounts for four types of accounts on one of the main social media platforms, Instagram. A private account, a public figure or celebrity account, a company account, and an organization account.
a) In the first month, the researcher will post the same content from each account daily. Weekly, the researcher will buy social media services such as likes, comments, shares, and views. These bought services will be rotated on each of the four accounts weekly. The researcher will combine observational research and true experiments to gather data. The data collected will be analyzed after three months. Based on the data and experiment that was carried out, the researcher will develop a new metric for measuring “bought social media services.”
b) This new metric will be applied to a new true experiment, which will be similar to the first one that was launched. The new experiment will be observed for another three months. After which the researcher will once again analyze the data and present the findings.
3.2 Research Graphic
In the past three decades, multiple types of research have been conducted to address a given problem, and in some instances, the outcomes have been the same or different across similar subjects. The difference, however, can be attributed to the multiple research designs which propose the overall strategy suitable for integration in a given study to address the problem under scrutiny adequately. The primary function of research design, therefore, is to solve the study problem logically using the evidence obtained and ensure that it is as unambiguous as possible (De Vaus & De Vaus, 2001). The paper will, however, not discuss all the available designs but instead will concentrate on the descriptive research and its components. A descriptive research design is suitable when it comes to an understanding as well as describing the characteristics which the population under study portrays (Shuttleworth, 2008). The paper will, as a result, compare and contrast components of descriptive research design such as observational, correlation, developmental, and survey studies. At the end of the latter, it will propose the best model that can be used in the study to be carried out and why it is appropriate for such.
3.3 Research Design
As an alternative of depending on anyone’s precise approach in developing this strategy, the researcher would like to work with a hybrid strategy that comprises the methods of the case study and investigational methods with reverence to the specific location’s circumstances. This will help the researcher in managing numerous difficulties associated with different locations in various ways. As such, the researcher can be guaranteed of extensive coverage of the project plan. It is always good to have multiple methods tried out on one particular product for evaluation. Thus, different methods of evaluation could have dissimilar outcomes, and this gives the researcher a choice to gauge the project professionally.
3.3.1 Developmental Study Design
The primary purpose of the developmental research design is to assess changes believed to have occurred within time. In other words, this design seeks to create knowledge from the existing data which have been systematically derived from a given practice (Richey & Klein, 2005). The relationship under investigation is not causal, which is similar to both correlation and observational studies. On the contrary, it differs from the other designs (correlation and observation) since it involves collecting data from the practices of the subject, which have been a part of them for an extended period and, therefore, cannot be reversed. In this case, the outcome of the study may remain the same when conducted by different authors, which is not applicable in the other two designs. There are three methods of data collection for the case of developmental study design, and they include cross-sectional as it is the case with the observational studies, longitudinal and cross-sequential. Longitudinal assess changes among the subjects of study over an extended period, which can range from months to years to come up with conclusive results. Cross-sectional studies involve reducing the mortality time of the research by assessing different ages at the same time rather than using similar groups over a specified period. Cross-sequential studies, however, incorporate both, which makes it stand out since the other two involve conducting a study on the same group.
3.3.2 Survey Study Design
The survey study design, unlike the three designs discussed above, involves administering or interviewing a large number of people (Yin, 2017). In such cases, the questionnaires are used to collect data that are then analyzed for differences, similarities, and even trends. The design compares to observational study as already mentioned in the previous paragraphs but differ in how data is collected. It is similar to all other designs also because it does not manipulate the independent variables to get the desired results. Compared to the correlation study, both can be used to determine the relation of the different variables.
In comparison to developmental study, it involves asking the subjects to write down their past experiences, which may have changed how they behave and interact. However, when compared to the same, it is unique when it comes to gathering information since only the original data gets to be collected (Owen et al., 2015). It is also unbiased when it comes to representing the interest of the population, and lastly, it allows for the standardization of different measurements. The best design for the study, therefore, is the correlation model since it allows a comparison of two variables to determine the relationship that exists between them.
3.4 Content Analysis
A content analysis study was conducted. Content analysis is a research technique to analyze documents and text and tries to quantify the content. It is a flexible method, and it can be used on visual documents. It seeks predetermined categories and systematic characteristics. The disadvantage with the content analysis is that it might highlight things that are not that important from a theoretical perspective. The focus of this study was the posts made by one brand on Instagram. The method is broadly used in social media studies. Content analysis is also very transparent and easy to replicate (Bryman, 2012).
Every business operates a social media platform that is either controlled internally by the marketing team or IT department or through an outsourced social media agent or administrator, depending on the size of the entity. In this protocol, a platform is used to mean an interactive internet site that allows companies to interact with their consumers directly. On the other hand, Hit is a representative of the interaction between the target audience and the post on the social media platform. At the same time, a post is a promotional message that is uploaded by firms to their clients.
The return on engagement is a measure of payback or returns from an enterprise’s investment in interactions on various social media sites. This paper is a protocol for the administration and collection of necessary data to support the determination of whether the return on engagement is a relevant measure of social media engagement. Social media has become an important element in organizational communication strategies. For communication research, social media provide a gold mine for big data collection (Liang, & Zhu, 2017). According to Ackland (2013), data collection approaches have changed in recent years due to advanced technology. The authors cite social media as the newest method of data collection in social sciences.
Scholarly Technology Group developed the Social Feed manager tool for harvesting data on social media to support research and build archives (Farseev et al., 2015). The SFM tool provides support for writing social media to Web ARChive (WARC) files. Application of the WARCs is a strategy that closely aligns social media harvesting with web harvesting (Olmedilla, Martínez-Torres, & Toral, 2016). The current research discusses the common techniques applied in the retrieval of social media records from the application programming interface and writing to WARCs. Data harvesting can be used to collect different types of data, such as content and behavioral data, and network structured data.
Many social media platforms such as Twitter and Facebook provide APIS to allow users to retrieve social media records. Platforms such as Twitter use APIs called Twitter REST API and the Tumblr API (Russell, 2013). According to Lomborg & Bechmann (2014), API uses HTTP as the communications protocol to create access to records in readable formats. Social media harvesting from APIS offers advantages such as APIs, which are more stable. APIs creators understand that changing the API will break consumers of the API.
The API provides notifications to the users at any time that they are making changes. It is to notify the users of the possible changes and give the users a chance to interact with the changes before they are fully implemented. If the API fails, then it will not be implemented. The API provides social media records in formats that are intended for machine processing. The data harvesting and analyzing are done through machines, unlike data collected through surveys. The API can provide data that is not accessible through the website platform (Blazquez & Domenech, 2018). API provides a backdoor to the data collection team to access data that is not readily available on a social media platform like Twitter.
To simplify the data collection methods through interacting with social media APIs, the developers created API libraries. The API library uses a specific programming language to make it easier for the user to interact with the API by handling rate limiting, authentication, and HTTP communication low-level details. According to Oh (2012), the web scraping method can also be applied as an alternative to the usage of APIs.
3.5 Data Mining Protocol
Data harvesting on social media uses machine learning (Injadat, Salo, & Nassif, 2016). The data collection team must evaluate the type of data they need before they choose a mining method. The researcher has to use a developed algorithm suitable for acquiring data from various social media platforms. Data harvesting is done using computers, thus the need for clear objectives to avoid collecting unnecessary data (Xu & Zhu, 2016).
3.5.1 Analysis Protocol Plan
In the process of data analysis, modeling is the main process. Data modeling include various process, including reducing data sets dimensionality, employing data modeling techniques to data, and obtaining the results (Aggarwal, 2011). Depending on the type of data collected, two data modeling paradigms may be applied: supervised and unsupervised learning (Alfantoukh, & Durresi, 2014). Supervised will be applied in situations where the researcher’s problem, e.g., the data sets, have inputs or independent variables and outputs are the dependent variables, and the main goal is to use the dependent variables to infer the values of the dependent variables. The problem can further be categorized as a classification problem where the outputs are expressed as categories or regression problems. Unsupervised learning refers to problems where there are no relationships between the inputs and the outputs. The types of problems are categorized into clustering questions whose aim is to discover grouping in the data and to create association problems. The objective of the analysis method is to find rules that describe the behavior of the part of the unsupervised data (Oh, 2012).
Data harvesting strategy helped the researcher collect more data than in other methods. The harvesting methods involve gaining access to exclusive data. Unlike surveys that collect specific data, data harvesting on social media provide researchers with a pool of data that can be analyzed to provide the researcher with a bigger and deep content. Social media data harvesting is obstructive (Liang, & Zhu, 2017). The data harvesting method involves collecting data and analyzing social behavior without affecting it. It is more reliable and objective than surveys and experiments. Data harvesting in social media is an automatic process (Blazquez & Domenech, 2018). The process applies computers or machines to collect and code the data. The method saves time, effort, and monetary resources. Besides, the data harvesting approach can handle relatively large data compared to traditional methods (Zhu et al., 2011).
Data harvesting limitation is that it is hard to know the psychological and demographic variables of social media users. The problem with using computers to collect and code data is that computers are programmed, unlike a human who uses their brains, which can lead to making systematic errors in the process of collecting and analyzing (Liang, & Zhu, 2017). Wall & Stuart (2006), argues that the practice of data harvesting also faces the ethical issue of user consent and participant’s anonymity. Unlike in a survey where it is simple to acquire the user consent and collect data, the process of harvesting is faced by resistance as the researcher acquires access to the entire platform (Zimmer, 2010).
3.5.2 Administrative Information: General Administrative Information
The General Administrative Information for this research will be based on the entity’s IT department or marketing team, which should be provided by a leader. This because every company has a personal page on multiple platforms to ensure that it optimizes the ever-growing opportunities on the internet and the full contact information of every member of a team that has been assigned the role of monitoring a company’s social media platforms. In the case of outsourcing, third-party social media administrators will also be very helpful in providing necessary input with the permission of relevant authorities in the principal company. Moreover, the contact information of each member of the third-party social media administrating teams assigned to the principal firm should be provided by their team leader.
3.5.3 Total Opening and Operating Hours of the Internal and External Administrators
The opening hours and operating hours of those entrusted with the management of social media platforms within an organization, including the marketing team or IT department, are essential. Whenever the administration of a company’s social media sites has been outsourced to a third party, their opening and operating hours should be explicitly disclosed. The details of the duration of a business or external professional in a working week are relevant for the timely collection of input in the social media engagement. Whenever possible, the details within a third-party administrator’s entity should be made available to ensure a prompt response when they are requested.
3.5.4 Target Population and Data Collection System: Geographical Location of Company or Third-Party Social Media Administrator
The geographical location of a company is not necessary for this study because the access can be optimized online, given the nature of business depending on social media. To collect the information on the level of engagement on a firm’s platforms during a certain period, both internal and external administrators may be contacted over the same sites as a test of their effectiveness (Gupta & Karpe, 2018). However, for scheduled visits aimed at conducting physical verification of involvement, the location of a principal company and its social media administrator is essential.
3.6 Types of Social Media Engagement
As mentioned by Chang et al. (2017), there are three types of social media platforms that businesses use to communicate their promotional messages, including social networking, social reviews, and image-sharing sites. They are further separated into multiple private platforms, including Facebook, Twitter, Instagram, LinkedIn, Trip Advisor, Yelp, among others. Several key terms require definition in this report, for example, Post, Hit, and Platform (Powerpost, 2019). Social media engagement is a form of interaction with a business on different sites, which is expressed in the form of likes, shares, and comments.
In this paper, the researcher will analyze the interactions or reactions to business online posts and posters on the three most popular online platforms – Facebook, Twitter, and Instagram (Big Commerce Essentials, 2019). Engagement on Twitter is assessed in terms of retweets and followers. Retweets confirm the immediate reaction of consumers to a promotional message, while the number of followers shows their investment in acquiring more information about a business. On the other hand, interaction on Facebook can be measured by shares, likes, and followers. Shares and likes have the same significance as retweets on Twitter (Big Commerce Essentials, 2019).
Lastly, interaction on Instagram is measured by the number of likes and followers that react to business communication on the social media platform. Because of the proliferation of the internet in the US with reliable connectivity, the target population can be spread across all the demographics in the population (Dijkman et al., 2015). The data collection system in this research will include various tools, for instance, surveys, interviews, and focus groups (Guest, Namey, & Mitchell, 2013).
3.7 Methods Narrative
To effectively measure the ROI on social media, the best type of research method to be used would be a quantitative research design. Quantitative methods are important in generating numerical values that will be transformed into functional figures and statistics, which will be used to understand the profitability of social media investment. The quantitative method is applied in situations where the researcher aims to quantify individuals’ attitudes, opinions, and other variables that are used to generate results for generalization on a larger population (Kim, Sefcik & Bradway, 2017). Descriptive statistics will be used. According to Kim et al. (2017), the design is used because it is concerned with aspects such as explaining the relationship between two variables in term relation and where the study is performed.
Moreover, the descriptive design is recommended because Kim et al. (2017) states that it helps the researcher to understand the situation of the phenomena being studied. In most cases, the design is concerned with the collection of present or up to date information about the topic under investigation. The design is preferred because it allows the investigator to collect data that will be used to make adequate conclusions and informed recommendations (Kim et al., 2017). Lastly, the descriptive design allows the investigators to focus on the problem being studied and, in this case, measuring ROI on social media.
3.7.1 Focus Group
A focus group is composed of individual clients that come together to discuss a product before its development and is a reliable source of qualitative data (Cyr, 2019). A protocol is separated into three phases ranging from before, during, and after the unit has met (BC Student’s Affairs, 2019). Before the activity is conducted, the assessment plan should be filled by data collectors. Two focus group discussions will be developed in each of the four seasons over social media platforms for comparability purposes. A live interactive video on Facebook or Instagram should be planned as a focus group with the administrator and facilitator being the researchers. The latter should include questions on the issues related to the type and level of social media engagement and generate them along this line.
Focus groups are mainly formed in cases of research as they have evident suitability in research. They also allow investigation of ambiguous issues and collection of large quantities of data. Besides, focus groups allow the researcher to get a lot of information in a shorter period as compared to one to one interviewing (Morgan, 1997). The focus groups operate at the development setting as well as the team setting, which requires individuals who are team players, cooperative and authentic. In data analysis, the focus group analyzes data differently as it usually yields the benefits and the challenges to be faced by the firm once a particular decision is made. It has more advantages compared to other forms of qualitative data analysis. It is distinct since it analyses data in an explorative and descriptive manner that brings about advantageous and concrete decisions. Focus group gives a chance to an individual so as they can express the strength of their view on a particular issue once provided with opposing ideas and opinions as well as suit those who are unable to pronounce their thoughts with ease. As a result, the members of a focus group have shared ideas, and hence, the decisions are reached at a consensus (O”h Eocha et al., 2010).
The focus group for this research will comprise stakeholders from various companies that are beneficial to the research. The basic economic principle states that resources are limited, whereas the needs are unlimited. As a result, job opportunities are few while the population seeking employment is diverse. This, therefore, means that interviews have to be carried out to sift the qualified from the unqualified. Identify the skills that determine whether the individual will get a pass or fail in the interview. Persons seeking managerial positions are supposed to have soft skills and the ability to express themselves to the interviewers. However, interviews are also used in research, which applies the face-to-face question-answer technique. As such, this paper will serve to analyze and summarize interviews of the workplace as well as those involved in research, focus groups, and narrative data analysis.
The priorities of who should be preferred will be decided before commencing the interviewing process, as this will assist in avoiding the issue of consensus and dissent that may arise. The interviewee will have moderate management skills. Thus, the interviewee will possess the ability to delegate openly and practically. Furthermore, during interviews, the interviewers will pay a lot of concern to all problems that may arise before, during, and after any Focus group session. The researcher, alongside the interviewer, will observe the reliability of the interview and also observe unrealistic answers to questions. As such, the interviewer will portray the character of multitasking as managers have to perform more than one task at the same time handling employee’s performance reviews and even serving customers. Therefore, the focus group involved in this research should give a chance to accountable candidates, with strong decision-making skills and those who have creative and innovative ideas as they can solve problems experienced by the firm in a conventional way.
Organizations use focus groups in research and to determine the kind of employees they need. The priorities of who should be preferred are given before interviewing so as they can avoid the issue of consensus and dissent that may arise. They usually ask for behavior and competency skills such as team building and team management, negotiation, leadership, and adaptability. The interviewee should have some management skills such as the ability to delegate openly and practically, how they played their roles and responsibilities in a particular project, and explain why it worked. Furthermore, they should have the enthusiasm to show the benefits such as on expenditure, quality of services, their efficiency in the roles that they partake, and the customer satisfaction that will be derived from them.
Nonetheless, during interviews, the interviewers have some spots they pay a lot of concern. They observe the reliability of the interviewee, especially when a question is posed, and they blame other examples managers in a firm who don’t want to be held accountable for mistakes done by the junior employees. They also observe whether the interviewee is a team leader as it is a responsibility of a manager to inspire, motivate, and mobilize the employees to conduct their duties. Also, the interviewers observe the flexibility of the person since a manager should be open to new ideas, foster them, and if they have a positive impact on the production, they should be executed. Besides, the interviewee must have comprehensive knowledge of what the employer’s needs and challenges to market themselves and show that they are fit for the position (Straight to the Top, 2015).
Moreover, one should avoid giving unrealistic answers to questions. Also, they should portray the character of multitasking as managers have to perform more than one task at the same time handling employee’s performance reviews and even serving customers. This is a vital character that a person in a management position should possess. For management to be perfect, the focus group involved should give a chance to accountable candidates, with strong decision-making skills and those who have creative and innovative ideas as they can solve problems experienced by the firm in a conventional way.
Narrative data analysis involves diverse approaches to data collection as well as analysis. Narrative data is obtained from many, few, or single individuals and in many forms inclusive of responses to open-ended questions during a survey, notes derived from the field, published reports, submitted questionnaires, and even testimonials. For an individual to provide an analysis, one has to conduct a series of activities. To begin with, the individual has to familiarize themselves with the data, which will involve an understanding of the data, checking whether the methods of data collection are vague and hence determine if the information is biased. Secondly, investing a lot of time and effort will assist in giving the analysis more credit and merit. Also, one should provide the limitations that they feel will cause a decline in the quality of the data. One should recheck the solution needed and thus come up with some sample questions whose answers will result in the solution required. The categorization of data involves bringing together the ideas and concepts that are similar in the same group. Also, one can categorize their data by ciphering it or giving it indices and importantly assess the meaning of the codes. It may also involve the identification of the patterns and connections or relationships, especially between the ones that are directly and inversely proportional or when one results in another to create a sequence of the events with time.
Furthermore, summarizing the information collected from the field or obtained from secondary sources requires compiling all the data about that topic of study and taking the responses that are given by people either in support or oppose. Interpretation of data comes immediately after categorizing and assorting of data. For proper interpretation of data, one should make an abstract which will be used in writing the final report. Besides, the researcher should involve others to help in the analysis and interpretation of data. The researcher should also avoid inferring or generalization.
3.7.2 Interviews
Several interview questions may be fronted to consumers through social media platforms during live interaction (Salmons, 2015). A firm should ask for permission from the clients to optimize the level of participation. Besides, a company should conduct several phone interviews upon prior approval from consumers. After conducting online questioning, the administrator should request the respondents’ phone contact information, and when it is convenient to conduct a follow-up interview. Several questions may be fronted to determine the relevance of return on engagement as a social media metric, including:
a) Which social media platform do you use most and do you interact with a company’s promotional messages there (Kostyra et al., 2016)?
b) Do you consume product information to support your online shopping activities through social media (Hajli, 2014)?
c) According to you, is social media platforms a reliable delivery channel for such marketing information (Jiang, Yang, and Jun 2013)?
d) How frequently do you interact with a specific company’s promotional messages on social media platforms – Facebook, Twitter, and Instagram (Gopinath, Thomas, & Krishnamurthi, 2014)?
e) What is your general comment about a company’s platforms and their reliability as carriers of promotional messages over other conventional marketing channels (Killian & McManus, 2015)?
3.7.3 Surveys
Online surveys may be generated as a prompt in the business personal social media platforms aimed at requesting feedback from target consumers or users that interact with a company (Guest et al., 2013). Several questions may be asked, ranging from simple to complex ones aimed at collecting the necessary information to justify engagement as a social media metric. Simple surveys about different aspects of consumers’ experience while using platforms for the acceptance of promotional messages from companies will be conducted regularly (Mosteller, Donthu, & Eroglu, 2014). Two complex and detailed survey sessions will be scheduled during distinct seasons of a year, including winter, summer, autumn, and spring. It will ensure that the opinion that is derived is not seasonally biased with the help of the same social media platforms, which will be analyzed in terms of the number of interactions and the content of responses provided.
In the process of conducting a study on the relevance of return of engagement as a social media metric for business purposes, a plan or protocol for data collection and analysis is critical. Facebook, Instagram, and Twitter are the most popular social media platforms, and they will be relied on for information gathering in this research. Surveys, questionnaires, and focus groups are the tools that will be employed in the protocol to generate input. Primary data will be most useful in assessing the ROI on social media. The method will help the researcher to obtain first-hand information from the stakeholders who are involved in social media marketing about how it is prodding in different companies (Brannen, 2017). The main methods that will be used to collect information include the usage of questionnaires, surveys, and interviews with social media executives.
The primary data collected will later be analyzed quantitatively in the form of graphs, figures, percentages, and other statistics that will enhance the understanding of ROI on social media. The interviews and surveys will be designed to investigate ROI on social media in terms of revenues, the number of subscribers, customer service satisfaction, customer acquisition, lead generation, clicks, and contest entries, among other factors (Saura, Palos-Sánchez & Cerdá Suárez, 2017). Precisely, the interviews and surveys used will be based on different social media KPIs. In this case, the interview will focus on asking questions and collecting data on KPIs such as customers who are actively engaged, leads generated, traffic generated on the social media, impressions, followers or fans, profile visits, brand mentions, comments, and likes, and shares. The above variables fall under the four main KPIs that will be measured when deciding on the ROI of social media. The four KPIs that will be focused on include customer engagement, reach, leads, and conversions that have been obtained through the use of social media (Saura et al., 2017).
Secondary data methods will also be used to examine previous information about measuring ROI on social media. Official Statistics refers to the official or endorsed statistics that are collected by government agencies such as bureaus and its departments. Primarily, such statistics will be of great value to the study as it is easily available and spans over a long time. Technical reports will be used as one of the sources for data collection. Technical reports usually involve publications from the topic by experienced experts in the field. The advantage of technical reports is that they are usually published at the request of an organization, government, or institution to address a specific aspect of concern (Brannen, 2017). Scholarly journals contain original researches which have been published by experts in a given industry or field of study. Trade journals contain practical information about a given filed. The information coined in them is key as it is highly practical from an application point of view. The use of secondary data has many benefits that will contribute to the value of the study. For instance, the method saves on time since the researcher will aim to use mostly search engines and other readily available sources, thereby avoiding frequent travels from one region to another in search of study respondents. Secondly, collecting data through secondary methods is affordable because there are no transportation costs and other miscellaneous expenses such as printing overheads that are associated with primary methods.
3.8 Data Analysis
Regarding data analysis, a mixed-method approach will be used. Both quantitative and qualitative data analysis methods will be used to gain insights on the ROI of social media. The use of different methodologies is of great significance, particularly when the investigator wants to overcome the weakness in a certain data analysis method (McCusker & Gunaydin, 2015). Applying both quantitative and qualitative approaches at the same time makes the investigator achieve a more profound understanding relative to the situation under investigation as it aids the scholar in balancing the limitations on one approach with the advantages of the other method, thus increasing precision (McCusker & Gunaydin, 2015). All the data collected on variables and how they influence ROI on social media will be analyzed using regression analysis to establish the nature of strength between them.
3.8.1 Stakeholder Analysis
The research participants for the study will be advertising professionals employed by advertising companies. These experts specialize in customer services dealing with social media. The study will use qualitative data instead of quantitative. It will involve interactive and direct dialogues with the participants to attain a shared understanding, which will provide it with authenticity and validity (Aluri, Slevitch, & Larzelere, 2015). The firms selected for the study will be based on convenience sampling. The role of each participant in the study is to provide answers from their experiences of using the social media platform. The participants are expected to understand the social media engagement metric, as well as the relevance of the return on engagement (Luarn, Lin, & Chiu, 2015). The researcher will interview professionals employed by companies of different clientele, locations, and sizes to attain a broader scope of perspective.
The sample will include corporations with various capacities, from 15 to 900 workers. The advertising companies should understand the online related activities of the customers. By understanding the clientele, the firm will be able to direct consumers in a way that is in sync with the goals of the business (Muñoz & Towner, 2017). Motivations for the overall use of social media are social interaction, entertainment, and information sharing, as well as empowerment purposes (Payne, Storbacka & Frow, 2008). Beyond the motivation of the consumer to utilize social media, the participant needs to understand how to successfully create a social media campaign that communicates to consumers (Laroche, Habibi, & Richard, 2013). There are various ways of building a social media advertisement, and it depends on the business goals.
All the firms will be full-service, conventional advertising firms situated in America. The study will collect information on perspectives of businesses that regularly work on the social media campaign for different clients ranging from non-profit, business-to-business, and government clients. The answers and perspectives will be included in the data analysis to provide a well-rounded overview of every company (Huang & Benyoucef, 2013). Firms with over 250 workers will mostly require speaking with the people in the department handling social media projects for customers. Smaller companies juggle both social media campaigns and conventional advertising campaigns for their clients (Delerue, Kaplan, & Haenlein, 2012). Differences in size were the main components when selecting the agencies to conduct the data collection.
Professionals at the advertising establishments will be contacted through LinkedIn messaging and email to ask for permission to be part of the study. The email will be used to set the date and time of the interview. The advertising professionals who will be interviewed will include the CEO (Chief Executive Officer), Social Media Director, Analyst for Digital Insight, Analysts of Social Media, Accounts Manager, Interactive Account Administrator, and Interactive Production Director. This variety provides the researcher with a diverse viewpoint concerning measurements of successful advertisements on social media (Chua & Chang, 2016). The interviewee will be asked to state their title at the place of employment and asked to define social media for the investigator to establish if the participant is conversant with social media. The other questions will be about measuring social media for the clients.
Most of the advertising firms use cost-based criteria, qualitative assessments, and internet measures like clicks and page views to make decisions (Kaplan & Haenlein, 2010). Conventional brand metrics like qualitative assessments provide the measure of brand passive interest-only (Chang, Ku, & Chen, 2017). Presently, there is nearly close to real-time information captured of consumer responses to advertising, and it has various effects. Fragmentation of the audience and the proliferation of media makes the traditional methods of advertising less compelling (Drake et al., 2019). Advertisers can measure the efficiency of their efforts and concentrate on promotions only on the targeted consumers who are interested in particular categories, products, or brands. The ability to track the relationship between sales and advertising has grown to become sophisticated and precise (Gill, Sridhar, & Grewal, 2017). The advancements give diverse viewpoints about the measurement of the successful campaign.
The professionals working in advertising companies will make this study successful because they closely monitor social media to use it as an advertising tool. The advertisers monitor social media daily by communicating with the customers to maintain consumer loyalty (Knight, 2016). Social media needs constant monitoring compared to the traditional method of advertising because of two reasons: the fluctuation of the audience and two-way communication (Kaur et al., 2018).
Corporations must engage customers through conversations and transparency. When social media is not monitored appropriately, a firm’s brand may generate negative perspectives (Perrin & Anderson, 2019). Social media has the power of reaching new customers and audiences who might have been unavailable before because of the lack of a social media platform. Furthermore, social media offers advertisers a chance to promote business and interact with potential consumers leading to long term relationships. This research will be beneficial to the participants of the study, as it will assist in promoting further their business.
3.8.2 Research Participants and Roles Assigned
This research will generally involve participants such as long term and valued customers, the C-Suite of companies such as (CEO, CFO, CIO), the human resource manager, the marketing manager, the finance manager, and the IT manager. All social media activities must be aligned with the company’s brand’s vision and mission (Carr et al., 2015). All of these entities will coordinate to determine the monetary value of the company’s social media marketing (Sajid, 2016). Each of these participants will have a key role to play in ensuring the project is completed. The role of the HR manager will be specifying the company’s goals and assigning key resources to the project (Huotari et al., 2015). The marketing manager is responsible for business advertisement and brand awareness strategies (Bianchi & Andrews, 2015). Involving the personnel will be crucial in establishing the targeted customer segmentation as well as the techniques employed. Considering the social media is a digital platform, it would be necessary for involving the IT manager in the whole process. In this project, the IT manager will be responsible for extracting social media data and performing the relevant data analysis processes (He et al., 2015). On the other hand, the finance manager will be accountable for assigning a monetary value to the social media goals and performing the required financial calculations.
3.8.3 Organizations Collaboration
Apart from the research participants mentioned above, the project’s success will rely on collaboration with external organizations. Significantly, these organizations will support the data analytics processes and tracking of the social media details (Baesens et al., 2016). In this project, the researcher will engage Sprout Social Inc. and influencers from various platforms such as Facebook, Instagram, LinkedIn, and Youtube in extracting data and performing the relevant data analysis. Google Analytics will be utilized to track business campaigns and objectives on social media platforms. This organization will be involved through the use of the Google URL builder to track social media links and the business expectations of increasing brand awareness (Batrinca & Treleaven, 2015). Sprout Social Inc. will offer a social dashboard to enhance tracking of the scheduled social media data (Benthaus, Risius & Beck, 2016). Generally, the researcher will involve these organizations in the data mining process to track the essential traffic messages relating to the business’ operations.
Chapter 4: Analysis
4.1 Facebook
Facebook is one of the leading social media platforms in the world. Most multinational companies have invested in using this platform to promote their products and services to their target markets. Therefore, there are several metrics associated with this social networking platform that companies can use to compute their social media ROI (Evans, 2010). For instance, the engagement rate formula for Facebook is used to calculate the engagement rate for a post by taking the number of engaging users and dividing by the total reach of that post. The result is multiplied by 100 to convert it to a percentage. In Facebook, engagements exist in the form of shares, comments, reactions, and some clicks on videos and images. An example of an engagement rate for a Facebook post is as follows:
Total engagement for a given day is 3,000, and the total reach is 90,000.
Therefore, the engagement rate = (3000 ÷ 90000) × 100 = 3.33%
The engagement rate for Twitter would take the following form. For instance, a company posts product that ends up getting total likes of 10,000 and 5,000 comments. Therefore, the engagement rate is as follows:
10,000 + 5,000 = 15,000.
For the company to arrive at its engagement rate, the result is divided by the company’s followers on Twitter and converted into a percentage by multiplying with 100. If the company has 500,000 followers, its engagement rate takes the following form:
(15,000 ÷ 500,000) × 100 = 3% is the engagement rate.
The engagement rate of Instagram also takes the Twitter formula. For instance, if the company posts an image that attracts 20,000 total likes and 10,000 comments, the engagement rate takes the following form:
20,000 + 10,000 = 30,000
The result is then divided by the company’s total Instagram followers and then converted to a percentage by multiplying by 100 as follows:
Suppose the company has 300,000 Instagram followers, then the engagement rate takes the following shape
30,000 ÷ 300,000 × 100 = 10
Therefore, the engagement rate = 10%
From the above analysis, a company can, therefore, calculate their social media Return on Investment as follows:
Profit ÷ Investment × 100 equals social media ROI %
In this case, the profit refers to the sum of money that the company has earned as a result of conducting marketing through social media platforms. Investment refers to the sum of money that acts as the cost of all social media marketing initiatives (Hoffman & Fodor, 2010). For companies to successfully use this mathematical approach, they should closely look at the marketing funnel and track their audience journey. The first step is to establish how and where customers discovered the company’s product.
Figure 9. tracking ROI down the marketing funnel (Schneider, 2008)
For instance, if a company is running two simultaneous CPC (Cost Per Click) initiatives in advertising its recent post on both Instagram and Facebook, the company can determine the number of users who have clicked on the ads of both of the platforms (Kelly, 2012). Therefore, the company can make use of a unique code snipper referred to as Facebook Pixel. Afterward, an individual can head over to Facebook Ads Manager to find out the number of customers who paid a visit to the company’s website. A simple example of social media ROI would take the following form.
Assuming that a book publishing company spent $10,000 on Facebook and Instagram ads, and got approximately 500 e-book downloads that enabled it to sell its products to at least 1500 customers, its social media ROI would take the following form (Kumar & Mirchandani, 2012). First, the cost of downloading each item would be $10, and each customer was an expense of $25. Considering that each customer bought $100 worth of products in the company’s brick and mortar store, the revenue would be calculated as follows
1,500 customers × $100 – $10,000 = $140,000
The company’s revenue is $150,000, and its profit is $140,000
4.1.1 Facebook Content Reach
Calculating content reach for Facebook is achieved in several ways. Brand reach, which refers to the number of users, is added to the story, which includes likes, shares, and comments (McCann & Barlow, 2015). The result is multiplied by the total number of users who engaged with likes, comments, and shares of the post.
4.1.2 Weekly Page Engagement Rate
The engagement rate of Facebook can also be calculated weekly. Therefore, this engagement rate equals the total number of likes, comments, and shares in a particular week divided by the total number of users during that particular day (Mochon et al., 2017).
4.1.3 Weekly Community Interaction Rate
Facebook’s interaction rate can also be computed at the community level. Therefore, the weekly interaction rate equals the total number of all the likes, comments, and shares by others to the posts made by the customers on Facebook divided by the total number of users of the brands’ page in that week.
4.2 Twitter
The engagement rate of Twitter can be calculated by taking the total number of replies and retweets and dividing it by the total number of followers on a particular day. The result is then converted to a percentage by multiplying it by 100 (Mochon et al., 2017). The average tweet engagement rate can also be computed by taking the total number of replies and retweets and dividing it by the total number of tweets on a particular day. The result is then divided by the total number of followers and then multiplied by 100 to convert it into a percentage (Powell et al., 2011). Twitter content reach is calculated by taking the brand reach, which is also the same as the followers of the company, then adding it to the shares. The result is multiplied by the followers who have retweeted. Lastly, the weekly community interaction rate for Twitter is computed by taking the total of all the retweets, quotes, and replies (Sen et al., 2018). The result is divided by the total number of followers of the company.
4.3 Return on Investment on Social Media
Social media ROI can be computed by subtracting the cost of investments from the earnings or gains or benefits of the company. In this case, the earnings or gains refer to the sales, leads, and inquiries from the investment. The cost of investment refers to the amount of money incurred as a result of conducting a social media campaign (Crumpton, 2014). This figure is also closely related to the agency commission, retainers’ fees, Cost per Click, resource cost, or any other amount of money incurred as a result of undertaking the campaign.
4.4 How Companies Can Measure Social Media ROI
4.2.1 Have Clear Objectives
Brand awareness that is created by social media paid or organic likes is always valuable but not significant enough to increase a company’s profitability. According to research from Altimeter, only 34% of companies are confident that their social media strategies are parallel to their revenue outcomes. As such, for a company to achieve value from paid and organic likes, it needs social media objectives that complement departmental goals and business (Crumpton, 2014). For instance, the company’s social media objectives should be based on consumer experience, creation of brand awareness, security and risk mitigation, and business conversations like lead generation.
4.2.2 Set Smart Objectives
The objectives of a company must represent how social media will aid the company to realize its goals. For instance, business conversations present a company’s sales team with quality leads through social media. Companies can achieve business conversations by driving 30 email sign-ups monthly. Secondly, consumer experiences will turn consumers into loyal brand advocates. Such can be achieved through the reduction of average timely responses on social media to one hour. Moreover, a company can achieve brand awareness by investing in advertising its new products before it is officially launched, thus taking attention away from its competitors.
4.2.3 Tracking the Appropriate Performance Metric
Companies require social media metrics such as comments, shares, and likes to evaluate whether they have realized their returns on paid and organic reach investments. The social media metrics are usually used to gauge the general health of a company’s social presence, determine the content that resonates with their consumers, and place itself against its competitors.
4.3 Determining Investment on Social Media
Companies must evaluate and determine their scope on social media investments to understand whether they are gaining positive returns. As such, a company’s management can evaluate their returns using four key factors. Firstly, management must determine the cost of social media platforms and tools. Most social media platforms are free to use, but companies can pay for a premium version of the social media management platform (Felix et al., 2017). As such, the company can measure its return on investments by taking the monthly cost of the social media platform and divide it by two. Secondly, a company can determine the amount it spends on social media through the budget allocated to social media expenditure. The cost of social media Ads is usually easy to track as it is recorded on the platform’s Ad dashboard. However, if a company is using multiple networks, it can employ the use of tools like Hootsuite Ads to manage each campaign and measure returns on investment from a single platform.
Moreover, the time spent by employees responsible for creating, promoting, and running ads can be added up and used to conclude the return on investment of a campaign. As such, the business can be able to calculate the amount of time the business spends on social media monthly and annually. Finally, the expenditure on content creation can also be used to determine a return on investments. As such, the company can determine the expenditure on content creation by determining how much a writer makes in an hour in case it employed blogger services.
4.4 Selection of Social Media Platform
Instagram was chosen for this study because of its varying qualities and different options to make a post. Because of the visuality of the platform, it is an interesting place to do marketing. Instagram keeps on growing each year, and it is constantly gaining new users of different ages. Based on “We Are Social and Hootsuite Digital Report,” the main users of Instagram are between 18 to 34 years old (We Are Social, 2018). The chosen brand products are suitable for all ages, but it could be assumed that younger consumers are the majority of the consumers who purchase them.
4.5 Data Collection
The analysis concentrated on a time limit of six months, January-June, 2019. All the data was collected by saving all the posts posted by the selected brand on Instagram over this period. The data that was collected is visible to everyone, even without logging into Instagram. All the posts were screenshot, and the numbers of likes and comments were recorded from the same day. This because on Instagram, there is always the possibility that more people like the post, or then something related to the posts, can be modified or deleted. All the post related information was collected, registered, analyzed, and finally put to Microsoft Excel. The information regarding the pictures or videos, i.e., date, message of the post text, number of likes, and number of comments were collected during that period. See tables 1, 5, and 6 for how the data was coded. The coding scheme is partly based on the literature reviews regarding Instagram postings and engagement levels.
4.6 Findings of Content Analysis
The findings of the content analysis are presented based on the research questions. First presented is the profile of the brand and then the different characteristics of the posts. The second part is more focused on engagement and how the posts collected likes, comments, or views. The total numbers represent all the findings, but because there were so many reposts, these can also be seen in their separate category. Six of the posts with the highest engagement rates are also presented in a different table.
For this study, four brands were chosen based on their activity on social media platforms and especially on Instagram. The brands are Nike and Starbucks. Starbucks is a start-up, and it offers an e-commerce platform for individuals to buy or sell various products from the precious metal industry. The products can be found from several stores and online purchase sites.
4.6.1 Case Study
In the 21st century, it is significant for companies to take advantage of the opportunities available in digital marketing. Social media marketing is essential for any company looking to make more profits and attract a large multitude of consumers. Social media marketing is essential for companies in the contemporary world due to several reasons. For instance, it plays a vital role in building awareness of a business’ brand. It also plays an integral role in communicating authority in a competitive environment. Some of the companies that have invested in social media marketing are Nike, Starbucks, and Coca Cola.
4.6.1.1 Nike
Nike is the world’s leading sportswear and accessories manufacturer and has been able to effectively make use of social media to reach out to its customers. In recent years, this company has proven that it entirely comprehends how to tap into the mood of its target market, while simultaneously accomplishing its wider objective of inspiring individuals through the power of sport. A good example of its social media presence is the ‘Play for the World’ campaign (Piskorski & Johnson, 2012). The campaign was showcased across its social media channels, which emphasized the importance of social distance during the current COVID-19 pandemic. Generally, Nike uses social media to communicate with the community and the public at large.
4.6.1.1.1 How Nike Uses Social Media
Nike is regarded as one of the most dominant sports brands globally, with an approximate of $16 billion worth (Constantinides, 2014). Like most companies, Nike has a unique social media presence with separate Facebook pages for each of its product lines. These product lines include skateboarding, golf, and Nike+ Run Club. It also manages the other two football pages specifically for the American market and other markets. In particular, Nike Football boasts of more social media likes than its official account. The football version account has 44 million likes, whereas its main account has 32 million likes (Kwon & Sung, 2011). These figures are more than the Facebook page of Nike’s closest competitor Adidas, which boasts of approximately 35 million likes. Also, Nike has a dedicated page for its specific product varieties like Air Force 1 and Air Max. It also has a unique account to focus on women’s products and other female professional athletes. A significant observation of its Facebook use shows that Nike uses the platform moderately for its organic social production. Also, it makes use of the social platform for significant platforms like ‘Awaken the phantom campaign.’ As it makes its leading athletes and their accomplishments at the center of their platform, posts still play a significant role in improving the company’s social media engagement through likes and comments. Nike also has the fourth leading social media following on Instagram, with close to 82.6 million followers, according to Social Blade (Piskorski & Johnson, 2012). It uses Instagram more often than it uses Facebook.
4.6.1.1.2 Nike’s Social Media Return on Investment
Nike has significantly profited from its online and social media presence. Generally, Nike often invests about $39 billion in its marketing strategies, and part of this constitutes its social media strategy (Piskorski & Johnson, 2012b). The company has observed several instances of Return on Investment from its social media strategies. For instance, the controversial Colin Kaepernick Ad Campaign was an example of a context where the company saw a rise in profits. Colin Kaepernick is a former NFL quarterback who made headlines for protesting police brutality in the United States. Kaepernick is known for his silent demonstration against police brutality by kneeling during National Anthems. He posted an image of himself addressing such vices in the United States, which contributed a lot to Nike’s profitability. Nike saw a 1,400% rise in social media activity, which led to $6 billion in sales.
Another instance of Nike’s social media ROI is based on Tiger Wood’s Nike sponsorship. Nike’s media value gauges the weighted impact of social media aspects, such as likes, comments, retweets, and shares (Piskorski & Johnson, 2012b). Generally, media values help a company to forecast the commodities that can generate significant social media ROI. For instance, the table below shows the social media value of Nike’s golf apparel line. The Tiger Woods’ Nike sponsorship is a perfect depiction of Nike’s social media ROI. During the 2019 Masters for Nike after Woods’ victory, the company gained at least $22.5 million from Woods’ on-camera exposure (Do et al., 2015).
4.6.1.1.3 Conclusion Case Study Nike
Generally, social media marketing refers to the application of social networking platforms to conduct a promotional initiative of a product or service. Companies are continually investing in social media platforms to ensure that they remain visible in their marketing initiatives. Also, in the contemporary world, social media platforms play a significant role in ensuring brands remain visible. For instance, social media is essential for companies because they are to communicate and engage with their customers. Social media ROI refers to the total value of all social media actions a company undertakes. According to Facebook, if an individual views an ad for a particular page and also likes the page, it is rendered as a paid like since it took place with one day of viewing the ad. However, organic likes are the number of individuals who see and like a user’s content without paid distribution. Today, several companies have invested in social media marketing. Examples of such companies are Starbucks and Nike.
The latter is the world’s leading apparel and footwear company. It has a unique social media presence with separate Facebook pages for each of its product lines. It has also significantly benefited from its online use of platforms like Instagram, Facebook, YouTube, Twitter, and other social media platform. A perfect description of Nike’s social media ROI is the Tiger Woods sponsorship. The company’s media value gauges the weighted impact of social media aspects, such as likes, comments, retweets, and shares. From this, the company has gained at least $22.5 million from Woods’ on-camera exposure.
4.6.1.2 Starbucks
In 2018, the world’s leading coffee house won a silver IPA Effectiveness award. It is because the company had an exceptional social strategy. A year before, the company had declared that it intended to release its iconic ‘Unicorn Frappuccino’(Taecharungroj, 2017). That was a business move that promoted the trend of businesses intentionally creating foods and drinks that can be advertised on Instagram. The key to Starbucks’ social media success is based on this understanding that such products influence young people. As of 2020, the company enjoys a wide base of Instagram followers, which is close to 18.3 million. Its entire social media presence is encompassed by approximately 67 million followers, which is close to as many as Apple and Amazon combined (Taecharungroj, 2017). The world’s leading coffee house was established in 1971 and operated in more than 29,000 locations throughout the entire globe. The company has been successful in pushing their drinks on social media. The company frequently posts employees’ favorite drinks, customer preferences, and relatable gifts and memes.
4.6.1.3 Epoxa GmbH Co KG
Epoxa is an online platform that enables users to buy and sell coins, medals, precious metals, and other collectibles on an E-Auction platform in the formats, Buy now/Offer/Bid. Epoxa additionally offers a DM to EUR exchange service. This service enables individuals who are located far away from the regional Bundesbank locations (www.ezb.europa.eu) to have the possibility of exchanging their DM currency to Euros. The company enjoys high prices of the precious metal for the exchange services. The precious metals are valued for the products associated with the metals.
Among the precious metals are gold, palladium, platinum, rhodium, and silver. The precious metal industry considerably differs from other metals and minerals because of the high prices and special properties contained in the metals. The metals are rare in nature and involve complex and costly mining and refining processes, which contributes to their high prices. The precious metal industry has unique and valued properties such as corrosion resistance, high conductivity, catalytic features, and are now used in advanced technology. The Epoxa GmbH Co KG company uses common social media platforms for marketing their products. The company’s social media platforms were used to analyze the return on investment of social media.
4.6.1.3.1 Epoxa GmbH Co KG Brand Profile on Instagram
As shown in figure 9., Epoxa GmbH Co KG has 456 followers. The company has highlighted some hashtags on its profile and added their shopping link. They have also saved video clips to the stories, and one of them was about the company. The company has posted a total of 27 posts, and the first post was made on 04.09.2018. According to a company representative, this post was created without having a single follower to commence working on the company’s social Media Image.
Figure 10. EPOXA GMBH CO KG Instagram profile
EPOXA GMBH CO KG has a main global Instagram page and many other ones for different countries. For this study, the company’s global page was selected because it was the most active of all the profile pages. They have made 27 posts in six months and re-posted 26 posts during that time. Figure 10 shows five of the company’s posts.
Figure 11. Five EPOXA GMBH CO KG posts
4.6.1.3.2 Type of Posts that Epoxa GmbH Co KG Post on Instagram
Table 1. summarizes the main characteristics of the posts. When analyzing the data in the table, some interesting differences in the features can be found. The analysis of the results will be mainly structured according to the variables presented in the table.
EPOXA GMBH CO KG – Content Categorisation
EPOXA GMBH CO KG EPOXA GMBH CO KG Repost Total
Number of posts 35 26 61
Text length average 20,8 words per post 41,8 words per post
Media type
Picture 28 22 50
Video 7 4 11
Audio Included (Yes) 3 4 7
Picture content type
Humoristic 2 1 3
Relaxed 1 4 5
Action/ Sport 0 0 0
Excited/ Party 11 0 11
Neutral 22 21 43
Picture content place
Beach 1 0 1
City 0 0 0
Home 0 5 5
Nature 1 2 3
Studio/ Other 34 19 53
Main content of the picture
Product 26 19 45
Face/Person/People 7 7 14
Nature 0 0 0
Other 3 0 3
Text content
Humoristic 2 0 2
Basic product info 6 9 15
Product purpose info 1 6 7
Excited 11 6 17
Neutral/ Other 15 5 20
Message style
Help to get to know products 8 12 20
Entertain/Neutral 7 5 12
Educate on practise/ trends 1 5 6
Convince to buy 12 2 14
Challenge to take part 7 2 9
Visibility of Brand products 32 24 56
Total number of Hashtags 30 110 140
Average number of Hashtags 0,86 per post 4,23 per post 2,29
Total number of Mentions 59 96 155
Total number of Tagging 3 0 3
Average number of Emojis 1,9 per post 2,8 per post 2,35 Av.
1) The number is affected by EPOXA GMBH CO KG post and re-post that has two pictures in one post The bolded numbers are the most interesting
Table 1. Characteristics of the posts
4.6.1.3.3 Pictures and Videos
EPOXA GMBH CO KG has made 27 posts containing 50 pictures and 11 videos between 1st July and 31st December 2018. The audio was included only in some of the videos. When analyzing the data in table 1, it can be seen that in the posts, the products stand out. Many of the pictures were simple and only showed the products, and therefore, they were categorized as neutral. In some of their posts, they have taken into consideration the holidays and different times of the year. There were Christmas, Thanksgiving, and Halloween themed posts. The pictures could not be placed in any specific physical place. Besides, the pictures included background themes, so these posts were classified as a studio. From the picture posts, almost half were reposts. When analyzing the pictures, it was noticeable that four pictures were reused with a different text. From the four, two of them were identical, while the other two were similar. One of the identical pictures was a repost, and this was posted twice with just a minor change in the text. The company also had one post with more than one picture and one repost with three different pictures in one post. Pictures were the most used post type.
The company had posted seven videos during the six months and reposted four videos. Four of the company’s videos were about the limited-edition line, and they were done as co-operation with an influencer. One of the videos was a commercial-type video, and all the reposted videos were tutorials. All the reposts contained audio, and only two of the company’s video posts did not have audio.
4.6.1.3.4 Post Texts
When analyzing the post texts, the message lengths varied a lot. EPOXA GMBH CO KG had an average of 20.8 words per post, and 41.8 words average per repost (See table 1.). The company had a lot of variations in their post texts. They mostly used the English language to shorten the words as well as using exclamation marks. They had seasonal texts and greetings to the followers, and different questions were asked to gather opinions; however, many of the texts were like small commercials where they promoted their products. They had a lot of offers like Black Friday and Cyber Monday and even two giveaways posts. EPOXA GMBH CO KG also had organized a fan’s meeting and was asking people to attend. In the texts, the company brought out the fragrance and the ingredients but did not open up the purposes of the ingredients. Most of the posts concentrated on promoting or getting the reader to buy the products.
The text content style was counted many times as neutral because there were not any specific characteristics in the text. They mainly mentioned the products and encouraged them to visit their web store to purchase the products. The second most used text content was excited, which means that they were trying to get the reader excited. This was used when promoting something like a new line or free shipping.
The company had some sort of question in nine of their posts. Two of them were the giveaways, and in the others, they were asking for a favorite scent or favorite fall product. They asked hair-related questions from the followers, and a few could be counted as rhetorical questions.
When analyzing the reposts, many of them mentioned the fragrance and brought out the ingredients. The reposts had more tutorial type text and told more about the use and the effects of the products. Hence, there is some difference between the text content of the original posts and the reposts. When the total number is studied (see table 1.), it can be noticed that most of the posts concentrated on the products and some basic product information. Among the company’s reposts, nine posts had some sort of question. These posts had questions like “Have you ever tried?” or a question started the posts, and then the related issue or topic was discussed.
4.6.2 Social Media Statistics
On Instagram, the company received one of the leading engagements on its photos. It comprised of at least 30% of its published content. According to the posting frequency chart below, the company saw a 34.8% increase in its followers, leading to an additional 23 million new enthusiasts (Piskorski & Johnson, 2012). It also uploaded 182 posts during the period, where 62% of the uploads were videos. Therefore, the chart below suggests that the company has an inconsistent pattern of posting on Instagram.
Figure 12. A chart showing posting frequency of Nike
Nike also makes use of Twitter to tackle customer queries, with 96.8% of these tweets being replied. Based on the reply rate of the company, it takes an average of 10 hours for it to respond to the customers (Piskorski & Johnson, 2012b). The company primarily replies between 12 am and 1 pm. It also suggests that the company mostly replied to its customers on Tuesdays and Saturdays. The chart below shows that Nike replied from 11 am to 1 pm, with the reply rate reaching a peak at noon. It also implies that the company provided customer support almost on all days of the week. It also shows that the company responded quickly to the customers on Saturdays and Sundays.
Figure 13. Nike’s average reply rate on Twitter
4.6.3 Visibility of Brand Products
EPOXA GMBH CO KG had products shown in 88.6% of the posts, and thus there were only four posts that did not have them. The company reposts had only two posts without showing the products, and these posts were concentrated around the influencers. Clearly, on the original posts of the products were almost the only thing showing in most of the pictures, but on the reposts, more variations were observed. From the products shown in the post pictures, 36 posts out of 61 posts had the limited-edition products. After the limited-edition line was launched, there were only five posts that were not about it.
4.6.3.1 Posting Timeframes
The studied posting time frames can be seen in table 2. and table 3. Table 2. presents the months, and it can be seen that; of the total 61 posts, EPOXA GMBH CO KG had made most of the posts during the last three months, October, November, and December. The reposts were done each month and most of them during August, September, and November. The videos were posted every month except in December.
Table 2. Post months
Monday was the day when most of the posts were posted, however posts were made daily.
Table 3. Post day
4.6.3.2 Hashtags of the Brand
Hashtags were used in nearly all posts and only a few posts had no one at the company. The mean was 0,77 hashtags for each post and 3,78 hashtags for reposting per post. Nine different hashtags were used by Epoxa GmbH Co KG
, representing 66 various hashtags. A total of 30 hashtags have been used by the company and 110 replacements have occurred over the six months.
mentioned other Instagram profiles in their posts with 16 different mentions in a total of 59 times, and the reposts included 40 different mentions in 96 times. This generates traffic between the profiles because it creates a secure link between the pages. Mostly the mentioned accounts were the profiles whose posts had been reposted and the places where the products were sold. The most mentioned one was @nicolequerriero (in co-operation with the limited-edition line) in the original posts, and it was also mentioned in the reposts several times. In the reposts, the most mentioned one was the @Epoxa GmbH Co KG _beauty (the EPOXA GMBH CO KG profile). With the tagging, EPOXA GMBH CO KG had tagged some of the influencers, and with the products, they used click shopping linking. In the original posts, there were only three persons tagged on two different posts, and in the reposts, there was one person tagged.
EPOXA GMBH CO KG co-operated with influencers. The limited-edition line that was presented during this study time frame was done in co-operation with an influencer. Also, the influencer had posted EPOXA GMBH CO KG related posts on her profile, and these were reposted on the EPOXA GMBH CO KG profile page. Only one of the reposted videos mentioned that the post was done as a co-operation with EPOXA GMBH CO KG. From the other reposts, it cannot be said if they were done based on liking the products or was it a co-operation.
4.6.4 Consumer Engagement
Table 5. presents the amounts of likes, comments, and views collected from the posts. The videos cannot be liked, so they were not accounted for in the engagement rate. The average engagement for a post is calculated from the average number of likes and comments that a post received.
Engagement in Instagram
EPOXA GMBH CO KG EPOXA GMBH CO KG Repost Total
Number of followers 134 000 134 000
Number of posts 35 26 61
Picture 28 22 50
Video 7 4 11
1) Total number of likes 51 614 43 338 94 952
%based on followers 38,52 % 32,34 % 70,86 %
2) Average/divided with the number of posts 1843 1970 1899
%based on followers 1,37 % 1,47 % 1,42 %
Total number of comments 2717 999 3716
%based on followers 2,03 % 0,75 % 2,78 %
Average/divided with number of posts 77,63 38,42 60,92
%based on followers 0,06 % 0,03 % 0,04 %
Average engagement for a post 1,44 % 1,50 % 1,46 %
3) The highest engagement post 6,33 % 3,05 %
Figure 11. Average egagement rate 2018 1,55 % 1,55 %
Compared to figure 11. Eight posts over the average Seven posts over the average
%based on the number of posts 32 % 31,81 %
Total number of views 29 384 56 379 85 763
%based on followers 21,93 % 42,07 % 64 %
1) Only pictures can be liked
2) Only pictures are taken into consideration
3) ((Comments + Likes)/Followers*100)
Table 5. Engagement in Instagram
4.6.4.1 Earned Likes and Comments
Based on table 5. it can be noticed that EPOXA GMBH CO KG posts received a total of 94,952 likes. When total likes were divided by the total number of posts, a post received 1,899 likes on average. When calculated based on the number of followers, it showed that 1.42% of all the followers liked a post. The company’s reposts received more likes on their posts than the original ones. When analyzing the comments, the original posts received more comments than the reposted ones. In total, EPOXA GMBH CO KG received 3,716 comments during the counted time, and on average, a post received 6,092 comments, which is 0.04% of the followers.
When analyzing the engagement rates for the posts, the reposts received more engagement on average. When the posts were analyzed separately, it was noticed that the best engagement rates were achieved by the original posts. The original had eight posts over the 2018 engagement rate, and the reposts had seven posts over the average (to see the calculation formula see fig. 10. Average Engagement Rate on Instagram (Komok, 2018)).
The video posts cannot be liked; however, they can be viewed and leave some comments. The company videos were watched 29,384 times, and the repost videos 56,379 times. When the number of followers was taken into consideration, 21.93% of the followers had seen an EPOXA GMBH CO KG video, and 42.07% had seen a reposted one. In total, 64% of the followers have seen some a video.
4.6.4.2 The Most Liked, Commented and Viewed Post
Figure 11. presents the most liked, commented, and viewed EPOXA GMBH CO KG original posts. The company’s most liked post gained 8,252 likes, and the most commented received 1,506 comments. The most liked had 6.16% of all followers, and the one with most comments was commented by 1.12% of all the followers. The most-watched video was watched 8,993 times, and it was viewed by 6.71% of all the followers. As seen in figure 11. all the posts were neutral, and the most liked and commented posts had the products shown in them. They represent the same line and have similar colors. The most liked one was informed of a new limited-edition line and about the product launch. With the most commented one, it can be assumed that the post text was changed at some point, and the original text encouraged followers to comment, and this way allowed entering the giveaway. The most-watched video was a teaser of the same limited-edition line and was published before the most liked picture.
Figure 14. EPOXA GMBH CO KG’s most liked and commented posts
With the EPOXA GMBH CO KG reposts, the most liked posts received 379 likes that were 3% of the followers (See fig.12). The most commented one received 73 comments, which represents 0.05% of the total followers. With the videos, the reposted one was the most-watched and was seen 17,932 times by 13.38% of the followers. The pictures in the reposts were again quite neutral and contained the products. The most liked picture was again of the same limited-edition line that collected likes and comments in the EPOXA GMBH CO KG original posts. The text on the most commented ones had extra information, and the most viewed video was a tutorial which contained extra details on the products.
Figure 12. EPOXA GMBH CO KG reposted most liked and commented posts
4.6.4.3 The Most Engaged Posts
Table 6. Presents the posts with the highest participation rates for all posts. The similarities between posts that most liked and comments can be seen from the table. Most important similarities are the braid areas in the table. Compared to the 2019 average, the highest participation rate is very high. There are also high rates in other top posts. There have been a large number of loves for the numbers 1 (post 379.) and 2 (post 287), yet no comment has been made to the number two. A large number of likes and comments were received in all the other top posts.
Between the top posts are many similarities. Most post texts were classified as enthusiastic or as basic product info texts. Only one does not contain any additional information. The photos are neutral or reflect a thrill or a party image. All posts contain the images of the products and the main contents of the products are them. The most interesting post was a launch post on the services of Epoxa GmbH Co KG; the second and the third were donations; the fourth was again post-post of the new LLP, the fifth repost was the profile received on the LLP line and the sixth repost was EPOXA GMBH CO KG cocoon line.
The most engaging posts based on the Engagement Rate 2019
1 2 3 4 5 6
Brand EPOXA GMBH CO KG EPOXA GMBH CO KG EPOXA GMBH CO KG EPOXA GMBH CO KG EPOXA GMBH CO KG -repost EPOXA GMBH CO KG -repost
Post number 379. 31. 9. 38. 21. 58.
Average level 2018 1,55 % 1,55 % 1,55 % 1,55 % 1,55 % 1,55 %
Engagement rate 6,33 % 5,21 % 3,50 % 3,24 % 3,05 % 3 %
Number of likes 8252 6979 3182 4273 4019 3368
Number of comments
229
0
1506
68
68
73
Text length/ words 14 59 11 19 3 22
Text content
Neutral/ Other
Excited
Excited Basic product info
Excited Basic product info
Text message style Entertain/
Neutral Challenging to
take part Challenging to
take part Help to know
products Entertain/
Neutral Help to know
products
Extra value/info Yes Yes Yes Yes No Yes
Picture content type Neutral Excited/ Party Excited/ Party Neutral Neutral Relaxed/ Spa
Picture content
place
Neutral
Studio/ Other
Studio/ Other
Studio/ Other
Studio/ Other
Home
Picture theme Product Giveaway Products Products Products Product
Is audio included No No No No No No
Visibility of brand
products
Yes
Yes
Yes
Yes
Yes
Yes
Main content of the picture
Product
Product
Product
Product
Product
Product
Main message of the
picture
Product
Other
Other
Product
Product
Product
Weekday of the post Wednesday Monday Thursday Monday Monday Thursday
Number of hashtags 1 1 0 1 4 2
Number of emojis 1 5 6 1 3 3
Table 6. The most engaging posts based on the engagement rate
Chapter 5: Findings
5.1 Fake Likes
Studies on Instagram explore user behaviour [11, 20]. Especially close to our work is Jang et al.’s analysis of Instagram but without a focus on fraud [12] where they note the lack of reciprocity when it comes to liking behaviour. There are multiple sources of fake likes, such as paid web-services or apps, trading platforms where a user participates in giving likes in exchange for likes, and bots that are triggered based on hashtags. Instagram also allows users to post videos and maintains its view count and like count 1. We assume that if a video has received likes, but has zero views, then the like instances are fake because they were generated without properly seeing the content. We capture 16,448, such like instances (information about the liker, post, and source user), and add it to FakeLike_data. Such fake likes instances could have been generated from any of the sources mentioned above. The researchers acknowledge that the fake like instances dataset can be much more comprehensive, and leaves it for future work.
a) One view count increased by one when a user watched the content for more than 3 seconds. Like count increases by one on an explicit like action by the user.
b) Random Like Instances (RandLike_data): It is hard to obtain an exact positive dataset of genuine likes. Therefore, instead, the researchers collect a much larger random set of like instances to draw a comparison with fake likes and to use as a negative class to build a machine learning model to identify fake likes. Since Instagram does not provide a direct way to sample random users or posts, we obtain a seed set of Instagram users and extract their follower and followee connections in a breadth-first-search manner. This gives the researchers a sample of 1 million Instagram users, from which a smaller subset of users is taken and their likes extracted from each of the posts. In this manner, the researchers obtained a dataset of 134,669 like instances in RandLike_data. Note that this sample is much larger (more than eight times) than the fake like instance dataset. Therefore, despite the noise, the researchers assumed that predominantly, the like instances in RandLike_data would be genuine. Although a noisy dataset is one of the current limitations with a clean negative dataset, the results showing differences between fake and other likes, and supervised learning based identification of fake likes would improve. The data collected was summarized in Table 1.
#likes #posts (p) #Likers (L) #Posters (S)
a.
FakeLike_data16,4489,932 9,301 7,822
RandLike_data 134,6691,717 47,233 738
Table 7. Dataset of Instagram like instances. We also collect meta- information of likers, posts, and posters.
The researcher enumerates the plausible reasons behind a user genuinely liking another user’s post. The researcher begins by defining a like instance — given a poster whose post p has been liked by liker, define the like instance as the tuple (p). A like instance is designed to contain post properties to ensure that a liker is evaluated based on individual posts liked. It cannot be assumed that if a single like generated by a liker is fake, then all her other likes are also fake. Next, the following set of hypothesis which shed light on genuinely garnering likes on Instagram is defined.
The researchers extract a user’s interest profile from the bio, and by converting the post image into relevant text using Densecap. Note that this metric is not commutative and therefore penalizes likers with a wide variety of interests, which is an indicator of suspicious behavior. It has been seen previously that genuine likers tend to like things related to a limited set of topics and are more discerning [2]. Therefore, the researchers consider topical affinity as one of the distinguishing features to identify fake liking engagement. Note that the value of following the activities of the users being followed. Therefore, a liker can also come across the liked post p if it is liked by one of the users who follow the posts. The researchers also consider such an instance to indicate a higher level of confidence in the genuineness of the like instance.
To test H1.1 and H1.2, the researcher studied the follower and follower-of-follower connections to the posters in the dataset of fake and random likes. The researcher found that genuine likers do indeed like their followees post more than fake likers do (KS-test =0.303, p < 0.001). For fake like engagements, there are significantly less proportion of likers which are followers of the poster. In case of fake engagements, only 16.8% of likers of a post are followers of the poster, as compared to a much higher fraction of 39.1% likers being followers in case of random like engagements (Figure 1(a). Similar observations in a two-hop network in Figure 1(b) are noticed. Only 2.8% of likers of a post in case of fake likes are follower-of-follower of the poster, as compared to 42.4% in case of random like engagement.
5.2 Interest Overlap
H2: A user will have a higher chance of genuinely liking posts if they share interests. To capture interest overlap between two Instagram users, the researcher first define their Interest Profile and the extent of overlap as Affinity.
Interest Profile: Given a user U, the researcher defines U’s interest profile as a set of topics (T1, T2, …, Tn), where these topics are inferred from U’s affinity between [0, 1], with higher affinity value indicating high topical matching. The researcher found that 60% of fake likers have an affinity value of 0.475, as compared to 0.58 affinity for the same proportion of random set of likers. There are similar observations for affinity calculated over the bio, and only text extracted from the post image. However, the researcher combines both of these to enhance the discriminative effect.
5.3 Liking Frequency
H3: A liker will genuinely like more than one post of the poster. The researcher observes that legitimate likers keep coming back to the same poster. The researcher also found that 90% of posters with fake likes get 7% repeated likers on their posts, as compared to the same fraction of posters with other likes getting 42% repeated likes.
5.4 Influential Poster
H4: A user will have a higher chance of genuinely liking a photo from an ‘influential’ user or a celebrity. It was observed that celebrities garner a higher number of likes on their content [7]. The researcher used the Instagram verification badge as a proxy for celebrity users. It was observed that in the dataset, only 1.9% of users were celebrities who got fake likes, as compared to 7.5% celebrities who got genuine likes, indicating that celebrities are more likely to attract a higher number of likes. Therefore, the researcher considered a poster being a celebrity as a feature to gauge the genuineness of a like.
5.5 Link Farming Hashtags to get Fake Likes
Hashtags have been shown to play an important role in Instagram bio and posts (p1, p2 …, pk ) in spreading the reach of posts and attracting more likes [20], therefore the researcher examines their role in fake liking behavior. Due to strict API restrictions, we were unable to collect the entire 2-hop network of all like instances in our dataset. The results reported here are 55% of the likes.
4w 2vec (i, j ) ∈ [0, 1] is the word2vec similarity between the two entities, i and j maintaining a ratio of roughly 1:8.
1.00
0.95
0.90
0.85
0.80
0.00.20.40.60.81.0
1.00
0.95
0.90
0.85
0.80
0.00.20.40.60.81.0
This proportion ensures that any machine learning model trained on such a dataset can perform well ‘in-the-wild’ where the ratio of likes would be highly imbalanced. Therefore, the researcher obtains the features as mentioned above from FakeLike_data and RandLike_data and train a supervised model on these features with fake likes as the positive class. The researcher experiments with different classification algorithms viz, i.e., Logistic Regression, Random Forest, SVM (RBF kernel), AdaBoost (with Random Forest as base initiator), and XGBoost. Simple feed-forward methods are also used, i.e., Proportion of Link Farming Hashtags and Proportion of Topical Hashtags.
Based on the reviews from the hashtag posters, fake like instances show a high proportion of link farming hashtags which can be used to solicit likes but show a relatively lower proportion of topical hashtags.
H5.1: A user is more likely to attract fake likes using link farming hashtags in their posts. Link farming hashtags, such as ‘like4like’, ‘like2follow’, can be used to elicit fake likes. Such hashtags are often used by a community of users who collude to inflate each other’s likes and followers. The researcher curates a list of 112 such hashtags five and finds that 20.8% posts with fake likes have at least one link farming hashtag as compared to 1.8% posts with random likes. It was seen that fake like instances tends to have a higher proportion of link farming hashtags.
5.6 Topical Hashtags
H5.2: A user with genuine likes will have topical hashtags in their posts. Another aspect of posting behavior is related to topical hashtags used in a post, instead of occasional (#throwback- Thursday, #ootd), trending (#mayweather) or link farming hashtags (#like4like). The researcher adopted a two-step process to detect topical hashtags. First, filtered out all link farming hashtags as well as popular non-topical hashtags. Next, segmented the hashtags and used Wikifier to see what proportion of hashtags was about the topic. The researcher found that fake like instances tend to have a lower proportion of topical hashtags (KS test = 0.342, p < 0.001).
5.7 Detecting Fake Likes
In addition to the features based on the hypotheses previously described, the researcher considered user-based attributes, viz, the volume of posts generated, the average number of posts per day [2], and profile completeness (presence of profile picture, name, and bio) [8]. Finally, computed the Chi-square values to rank features in the order of their efficacy in distinguishing between fake and other likes. Topical affinity scores high (16%) as do user-based features, particularly profile completeness (11%), and the fraction of topical hashtags (9%) [7].
5.7.1 Building a Classification Model
While the actual ratio of fake to genuine likes on Instagram is unknown, based on previous literature on spam detection [4], the list in goo.gl/UshiXk was shortlisted from a list of popular hashtags by two active Instagram users. Like the link farming hashtags, popular hashtags were curated by the same annotators and are available in the same URL. The researcher excludes the follower-of-follower from the feature list since the 2-hop network of the entire dataset could not be collected due to API restrictions and neural network–Multi-Layer Perceptron (MLP), which gives us the best performance. In all the experiments, the researcher performed 10-fold cross-validation using 80% of the dataset as training data and 20% for validation. For the MLP based model, the researcher used two hidden layers with 200 neurons. Both layers use the sigmoid activation function, and the output layer has a dropout of 0.2 to prevent overfitting.
Baseline. As a baseline, the researcher used Badri et al.’s method to detect fake likes on Facebook [2]. Their work focuses on the properties of the liker, without considering the relationship between a liker and a poster. As the source code was unavailable, the researcher implemented this method based on the features detailed in the paper. The authors propose a supervised method for the detection of fake likes based on the profile (length of biography, the lifespan of account, number of bidirectional connections), posting activity (average number and maximum posts per day, total posts and skewness of posting), page liking (category entropy of pages liked, the proportion of verified pages) and social attention (average number of likes and comments received) of the liker. The researcher discarded two features: (1) the proportion of shared photos and (2) the average number of shares received since there is no concept of sharing posts on Instagram, unlike Facebook. The authors experiment with multiple classifiers and find XGBoost to perform best. The researcher uses Precision, Recall, and Area under the ROC curve (AUC) to measure the performance of all models in detecting fake likes, with the latter being important in understanding the performance of a classifier trained on imbalanced data.
5.7.2 Experimental Results
The researcher observed that boosted trees, Adaboost, and XGBoost (also used in the baseline), perform well due to robustness against outliers. However, the researcher achieved the highest performance using the MLP with an average precision of 83% and a recall of 81% (AUC of 89%) in detecting fake likes. The results are in line with previous studies on fraud detection, where neural networks have shown competitive performance [21]. The researcher compared validation loss and training loss across training epochs and observed that they are close to each other, almost converging with increasing epochs. The researcher, therefore, concludes that the fake like detection system does not overfit and generalizes well to unseen likes. Table 2 summarizes the results.
The baseline model gives a precision of 61%, a recall of 69%. Compared to the baseline model, the researcher’s model obtained 83% precision and 81% recall to detect fake like instances. The strength of the method lies in effectively capturing the parameters which affect genuine liking behavior on Instagram. The researcher performs a manual investigation of 200 randomly sampled fake likes labeled genuine (false negatives) by the baseline. The researcher noticed that users give these likes with well-formed profiles who superficially appear to be genuine users.
Below are the results of various classifiers in differentiating between fake and random like instances across different feature types. The researcher report precision and recall in the detection of fake likes. The MLP model performs best.
ClassifierLogRegRandom ForestSVM(RBF)AdaboostXGBoostMLP
Feature TypePrec Rec AUC Prec Rec AUC Prec Rec AUC Prec Rec AUC Prec Rec AUC PrecRecAUC
H1: Network Effect 0.10.710.320.120.670.350.160.720.660.150.750.590.170.700.640.240.730.67
H2: Interest Overlap 0.360.50.490.380.540.560.540.590.730.560.540.720.600.590.790.680.680.80
H3: Liking Frequency 0.110.620.380.120.530.360.180.60.650.130.350.610.150.430.680.220.510.71
H4: Influential Poster 0.110.620.330.110.50.350.120.710.60.160.60.680.220.590.690.310.700.72
H5: Hashtag features 0.210.560.540.210.550.510.280.60.690.490.260.720.440.340.710.600.510.76
User based features 0.320.160.570.30.210.570.40.220.700.530.610.740.540.670.730.610.740.79
All Features 0.390.670.670.390.640.620.580.650.770.650.600.780.690.650.810.830.810.89
Table 8. Results of various classifiers in differentiating between fake and random like instances across different feature types.
On the other hand, the method classifies 78% of these likes as fake likes. The researcher finds that the affinity between the likers and posters in these like instances are lower than average. The researcher also noted the presence of link farming hashtags in the posts of these like instances. Therefore, the system used by the researcher can detect fake likes given by genuine-looking entities, as well, unlike current methods that entirely rely on the user properties to determine fake liking.
Error Analysis. To understand why the model is not able to detect 19% of the fake likes, the researcher randomly sampled 100 undetected fake likes and manually inspected them. The researcher found that in 27 fake like instances, the likers were followers of the poster, potentially leading the model to misclassify such like instances as genuine. It suggests that some posters have fake followers, and the fake likes are from such followers, something the researcher’s methodology was unable to capture. However, the approach can be modularly applied in a cascade, after detecting fake followers using previous techniques [6]. Furthermore, the researcher found that 61 likers had a high topical interest overlap with the posts they had liked. A more thorough analysis showed that this was happening due to a small set of interests (just one or two) of the liker, which results in high-affinity value.
Chapter 6: Discussion
The purpose of this thesis was to find out what kind of content is effective in driving engagement in the context of social media marketing. The study was made by analyzing hair care brand posts on Instagram. Two research questions were raised:
1. What kind of posts an international hair care brand posts on Instagram?
2. What kind of post creates consumer engagement?
The findings of the study are discussed below.
RQ 1) What kind of posts an international hair care brand posts on Instagram?
• The results indicate that these are the main characteristics:
• Mainly pictures, but there are few videos and a lot of reposts.
• Picture content was mainly neutral and mostly about products.
• The text style was relatively neutral with basic product information.
• Most of them were posted on Monday.
• Average of 229 hashtags and 235 emojis per post.
EPOXA GMBH CO KG used a lot of different types of posts. The company mainly used pictures, but they also utilized other options like reposting and videos. They were very active by using user-generated content (reposting), and it clearly shows that they follow what the other profiles post about them. By doing the reposting, they bring more meaning to their content and connect with their followers. This made the relationship between EPOXA GMBH CO KG and the consumers stronger. The reposting seemed to be working as they had, on average, a higher engagement rate than the original posts.
On Instagram, 60% of the users have said that they find new products on this social network (Business Instagram, 2019a). It was noticeable in this study that there were not that many variations in the post pictures, and many of them had the limited-edition products shown. EPOXA GMBH CO KG concentrated on just one line, instead of bringing out the whole product line. Most of the posts’ pictures also contained the products only. The reposts’ pictures were the most versatile when it came to picture content. In the EPOXA GMBH CO KG original post pictures, the shampoos and the conditioners were the only things that were shown, despite having other hair care products. Vakratsas and Ambler (1999) claim that advertising must have some mental effect, and according to them, especially the frequently purchased goods are in the consumer´s mind already containing conscious and unconscious memories of earlier product purchases and usages. One of the EPOXA GMBH CO KG main selling points and strength is the product fragrances. The company could use the strength in their posts more frequently. For example, by bringing out the fragrances in the post pictures more clearly. It should also be noticed that there were no pictures of healthy or damaged hair as an example of the suitability of the products. In the posting text, the good fragrances were mentioned in several times, and it was clear that the fragrances were liked.
According to Sharma (2018), Instagram users engage more on weekdays, especially on Tuesday and Thursday. Most of the EPOXA GMBH CO KG posts were on Monday. The company made use of emojis, hashtags, and mentioned their original posts many times. This was also noticed in the reposts.
RQ 2. What kind of post creates consumer engagement?
The top posts indicate that the characteristics that create engagement in a post are:
• Excited text (product launch, information about something, etc.) or basic product info text.
• The post contains extra information.
• Picture is neutral.
• Picture contains the product or products.
• Giveaways/Competitions work very well.
• Picture posts are still the main thing, but videos are watched.
Consumer engagement in the context of social media marketing is encouraging the consumers to comment, like, and share content. According to Sharma et al. (2017) the “Face- book Engagement Pyramid,” liking is the lowest level of engagement, next comes the reaction, then commenting, then sharing and finally the creation of new content. Based on the findings, EPOXA GMBH CO KG was able to engage with the consumers. Also, based on the figure by Sprout Social “Brand actions that prompt the consumer to purchase decision on social media” (Chaffey, 2019), the top things were being responsive, offering promotions, and providing educational content. EPOXA GMBH CO KG made a lot of different post types, and the top things could also be noticed in the company’s posts. The reposting indicates that EPOXA GMBH CO KG is responsive, and during the six months, they offered two giveaways and took into consideration Black Friday and Cyber Monday. They had many reposts with tutorials, and in one of the posts, the followers had the opportunity to ask questions. These were answered by the company’s own hair care professionals (also influencers on Instagram).
As mentioned, liking does not contain as much engagement as commenting. The most liked picture of all the pictures got 6.16% of the followers. The post was quite basic and neutral, but it was an informative post about the brand’s limited-edition line. In other words, Instagram was used to inform the followers about a new addition to the product line. The most commented post got 1.12% of the followers to comment. That does not seem much, but it should be remembered that commenting indicates higher engagement. This post was a giveaway, so that ought to have activated the followers to comment and act. The other giveaway had the second-highest overall engagement rate, but it received no comments. Still compared to the average engagement of a post, the giveaway posts’ engagement rates were high. Although it should be remembered that all the people liking or commenting on the picture, do not necessarily have to be followers. The most-watched video was a tutorial, and it was viewed by 13.38% of the followers, which is a high number. On average, the video posts were watched a fair amount of time but did not receive any comments. Still, it could be said that the video posts worked well by reaching the followers and gaining their interest.
A previous study by De Vries et al. (2012) focused on brand post, like position and the influence of positive or negative comments, suggests that the top position of a brand is important, brand fans are influenced by each other and that posting a contest had a significant effect on the number of likes. In the same study, it was suggested that providing information was not important for the engagement rate. Compared to this study, some of the same requirements are met. It can be said that the brand here is a top position brand because it is sold in many countries and many places. Another similarity was that giveaway posts received high engagement. The influence of brand fans on each other was not taken into consideration in this study, as the profiles of the likes or comments were not registered.
Other qualities that can increase the engagement were noticed but could not be proven. For instance, EPOXA GMBH CO KG used a link between platforms. They had posts that had “tap to buy” or other comments like “stock up today,” “check out,” and “link in bio to shop.” This should increase the activity on their page and generate traffic to the company-owned sites. Another quality was posting frequency. According to a study on Instagram posts by Tailwind, the posting frequency could increase the engagement rates if the posts were posted often (Christopher, 2017). Based on their study, one post in a day would be the right number. Also, the hashtags were used in almost all the posts. According to Sharma (2018), using at least one hashtag creates, on average, 12.6% more engagement. The effect of the hashtags was not shown in the numbers of this study. With the posting timeframe, there was no noticeable difference between the likes of each day, so in this study, the effect of the posting day could not be noticed. There was neither a noticeable difference among likes between the months, even if in some months the posts were made more frequently.
Chapter 7: Conclusion
Social media platforms offer brands an excellent way to be connected to actual and potential customers. Social media is an easy way for brands to reach people, to stay connected, and to build the brand image. However, it requires a lot of work and planning for the brands, which is not always easy. Raising awareness is a key issue for many brands. For the brands, it is not enough to be present in social media; there is a need to attract and create interest. Besides, the interest needs to last, and a relationship should be created between the consumer and the brand for the engagement rates to be good. Based on this study that concentrated on the Instagram platform, the posts should be considered based on the brand style and the consumers for the brand. A few of the main things that rise were that the posts that got the highest engagement levels were pictures which contained the products. The posts had a neutral or exiting text and contained some additional information for the consumers. Also, new product launches and giveaways create engagement. Based on this study, there is not just one certain type of post that leads to success, but the key factor for a brand is to be active and engage with the followers.
Social media is vital in the management of a company because it provides a global platform where an organization can operate. It increases brand awareness (Di Virgilio, 2018). Almost half of the global population is on the social media network. Therefore, the utilization of the platforms to market products has the potential to reach out to a broad range of potential buyers. Social media is important in the management of a company because it comes with new ideas that assist the leadership in adopting global thought patterns and practices of business (Agnihotri et al., 2016). The advent of the internet has presented endless possibilities in the marketing sphere. The internet has become a great determinant of consumer purchasing decisions. One of the major avenues in which brands have opportunities to interact with their consumers over the internet is through social media (Laroche, Habibi & Richard, 2013). Social media has revolutionized the way societies communicate and interact in addition to sharing information and conducting businesses. According to Payne, Storbacka, and Frow (2008), these opportunities have been maximized by the business sector, with business entities making the most of the interactivity of the internet through creating more personalized relationships with their consumers.
Globalizing the operations of a company optimizes its productivity and presence in the market. Social media provides an efficient and effective mode of communication between people from various global regions. The efficiency and effectiveness in the flow of information play an instrumental role in a way that leaders and managers run the affairs of their organizations (Ramanathan, Subramanian & Parrott, 2017). The global aspect of social media connectivity can be damaging to a business organization and its products. One negative customer review can result in the loss of a significant measure of market share. The more complaints are shared on the online platforms, the more the company’s brand suffers. Social media comes with the element of diminishing face-to-face communication skills of members of staff and the management team (Rodriguez, Peterson & Ajjan, 2015). Thus, overreliance on the networks makes salespersons unable to sell their products physically to potential buyers. Employees of an organization represent its brand both on the offline and online platforms of interaction. The negative demeanor of an entity’s senior management personnel could hurt its business (Ngai, Tao & Moon, 2015).
With technology marching in line with Moore’s law, more processing power and more memory became available, with developers offering more features to their users at lower prices. According to a study by Pew Research poll indicated that half of the Facebook users in the United States get their news from social media as opposed to visiting news media website or reading newspapers. This is nearly 30% of the United States population (Fiorito, 2015). By the year 2020, Influencer Marketing will become a billion-dollar industry. Experts estimate that by 2020, the industry will reach $ 10 billion in marketers’ investment. Influencer marketing is getting more and more money. Companies hope for a good ROI for their brand from good partnerships and, thus, a good ROE. However, to determine and evaluate relevant metrics and prepare a campaign successfully, companies need a metric that not only factors the social media aspect but rather a metric that combines the social media metrics with the sale cycle.
Return on Investment is a measure that indicates the efficiency of business investment (Epstein & Yuthas, 2017). By using this financial indicator, a firm calculates the profit gained or losses suffered on the specific investment on the expenditures incurred (Watson et al., 2016). Normally, ROI is expressed as a ratio or in percentage form. Business organizations rely on this economic measure to appraise the firm’s profitability or the efficiency of investments made (Costa, 2017). Due to the emergence of digital marketing strategies, social media provides a grand stage for companies to evaluate their accountability (Alves, Fernandes, & Raposo, 2016). This is because social media has emerged to become one of the most effective marketing channels for business corporations. Precisely, measuring ROI on social media involves determining the level of a company’s brand awareness among the targeted customer segmentation. When measuring the ROI on social media, the major factors considered are the resources, time, and effort committed to advertising the company’s operation on social media as a means of increasing brand awareness (Buhalis & Mamalakis, 2015). The process of ROI calculation on social media involves the identification of the monetary investments on social media and attaching a monetary value to social media goals. Calculating the returns entails tracking the organization’s goals and assigning a monetary value to each goal (McCann & Barlow, 2015). On the other hand, social media investment is calculated by figuring out the financial value of the time, social media tools, and media advertising factors. After generating these financial approximations, the basic formula of ROI is used to calculate the business efficiency on social media (Aluri, Slevitch & Larzelere, 2015).
This paper, therefore, seeks to explore consumer engagement through the lens of return on engagement to establish the validity of return on engagement on social media engagement. In the process of conducting a study on the relevance of return of engagement as a social media metric for business purposes, a plan and protocol for data collection and analysis are critical. Facebook, Instagram, and Twitter are the most popular social media platforms, and they will be relied on for information gathering in this research. Surveys, questionnaires, and focus groups are the tools that will be employed in the protocol to generate input. As a result, the proposed research design will be descriptive under the quantitative study approach. About data collection methods, the researcher plan to use both primary and secondary sources. Primary data will be sourced from the selected advertising companies through interviews.
Sample questions to be asked during the interview include:
i. How many are social media users visiting the site based on their locations?
ii. What is the level of engagement with social media users on the platform?
iii. How many leads have been generated since the platform was made operational?
iv. How many followers does the social media platform have?
v. What is the sharing rate of each post made on the social media platform?
vi. To what extent do clicks made by customers during their last engagement attrition lead to a purchase?
Secondary data will be obtained from published sources on the topic. The variables for the study will be analyzed using mixed methods, which include ROI on social media. ROI is the dependent variable and level of engagement, the number of social media users visiting the site, the number of leads generated and clicks made by customers from the independent variables for the study. The variables will be quantified using the software.
7.1 Limitations and Future Research on Paid vs. Organic Likes
A few limitations have been found while this study was conducted. These limitations came from accessing all the information, setting the boundaries of the study, from the right time frame, and the amount of data. The whole study is based on what can be accessed by everyone and not by the information that the business profile can provide. With the information from the business profile on Instagram, the study could have been more exact. One of the main difficulties was also creating boundaries for the pictures to be analyzed. Pictures normally say more than a thousand words, but now there needed to be
Some boundaries that were used to code the data. Another issue was the time frame. Many times, the trends influence the posts, and this study concentrated only on six months. Instagram also keeps constantly changing by adding new features, and these factors always have an impact. There are artificial purchased likes and fake pages that might corrupt the figures. It would have also been good to study different brands and compare their results to each other, to see if there would be major differences. Social media is so big that it is important to restrict the area of study because the options are so massive of what to research.
For future studies, it would be interesting to see how the posts have changed, as trends change, and Instagram makes new improvements. The comments and their content could be analyzed, and it would be interesting to see how the brands communicate with the followers.