Improving the impact of Facebook social media brand page posts

IMPROVING THE IMPACT OF

FACEBOOK SOCIAL MEDIA

BRAND PAGE POSTS

Master Thesis: Business Administration, Big Data & Business Analytics.

INTRODUCTION

Internet technologies have made new systems available to the public in the last decades. The

availability of the internet has given individuals the opportunity to use social media, such as Facebook or Twitter, to interact without the need for face-to-face meetings (Gruzd et al. 2011). Social media such as online communities is one of the systems everyone can use nowadays (Lu et al. 2010). Social media outlets constitute excellent vehicles for brands to create and maintain relationships with users (de Vries et al. 2012; Laroche et al. 2013). A specific way to do this is to use firm generated content (FGC) and to create brand fan pages on social networking sites (Jahn & Kunz 2012). Organizations are struggling to understand how social media content impacts their brands and customers (Booth & Matic, 2011). This thesis is focusing on is social impact through social media. Social impact is defined as the value created and experienced, both positive and negative judged against a benchmark of what the situation would have been without the proposed activity (Clifford, 2014). The question arises: How to measure the social impact of social media activities?

Brand managers place brand posts (containing videos, messages, quizzes, information, and other material) on brand fan pages. Brand managers have a goal with a community or brand page (Kumar, 2015). Customers can become fans of these brand fan pages, and subsequently indicate that they like the brand post or comment on it (de Vries et al. 2012) to network with other users (Kaplan & Haenlein, 2010).

Facebook (brand) pages are setup by all types of organizations, both profit and non-profit

organizations (Lovejoy & Saxton, 2012), and even by individuals (Preece & Maloney-Krichmar, 2005). As brand managers become more comfortable with using brand communities and brand pages for their integrated marketing communications, they have naturally turned their attention to questions regarding the success of these activities (Hoffman & Fodor, 2010).

Organizations have a corporate social responsibility (CSR) to measure. Furthermore,

non-governmental organizations (NGO’s) and social organizations have multiple stakeholders to account to, and several categories of interested parties who require accountability, and this makes the process of social impact assessment even more important (Ebrahim, 2010; Bagnoli & Megali, 2011; Lovejoy & Saxton, 2012; Nolan, 2014). The problem is, that measuring the impact and success of achieving the goals is difficult because of a lack of a standard for social impact assessment. The first objective of this thesis is to define how to measure the social impact of FGC social media posts.

Secondly, organizations and NGO’s likely want to improve their social impact. A number of methods are used to evaluate the proposed model to measure social impact and to find ways to improve the social impact. Are top of mind (what are people thinking about at a specific timeframe during the day, actuality news posts) posts improving social impact? Is the social impact increased when posts are linked to an external website with extra information? And by actively encourage interactivity and curiosity (for example by asking questions, or ending a post with and unending phrase) to boost social impact?

The goal of this thesis is first to define how to measure impact of social media posts, and second to evaluate methods to improve the impact of social media posts.

C

ONTENTS

Introduction ... 2

Theoretical Framework / Literature review ... 4

Impact ... 4

Social impact ... 4

Measure social impact in social media ... 4

Social media outcomes ... 5

Improve social impact in social media ... 10

Top of mind - Actuality ... 10

Extra information ... 10

Interactivity ... 10

Conceptual Framework and Hypotheses ... 11

Research hypothesis ... 11 Organization ... 13 Hypothesis testing ... 14 Results ... 14 Data overview ... 17 Discussion ... 18

Limitations & Further research ... 18

Acknowledgement ... 19

THEORETICAL FRAMEWORK / LITERATURE REVIEW

I

Impact is widely used in literature but rarely defined. A clear definition of impact where everyone in literature agrees on and how to measure social impact is not easily found in literature. Impact is by the dictionary defined as ''the force of impression of one thing on another'' (Webster's, 1986). Blankenburg (1995) defines impact as “long-term and sustainable changes introduced by a given intervention in the lives of beneficiaries”. Impact can be related either to the specific objectives of an intervention or to unanticipated changes caused by an intervention; such unanticipated changes may also occur in the lives of people not belonging to the beneficiary group. Impact can be either positive or negative.”

S

The type of impact we aiming for in social media is social impact. In the field of Social Psychology, social impact is defined as “any of the great variety of changes in physiological states and subjective feelings, motives and emotions, cognitions and beliefs, values and behaviour, that occur in an individual, human or animal, as a result of the real, implied, or imagined presence or actions of other individuals” (Latane, 1981). Social impact is also defined as “the consequences to human populations of any public or private actions that alter the ways in which people live, work, play, relate to one another, organize to meet their needs, and generally cope as members of society” (Burge & Vanclay, 1996; ICGP, 2003).

Maas & Liket (2011) and Grieco et al. (2015) did research on the question how to measure social impact. The lack of consensus on the definition of social impact and the best way to measure it hampers both the academic debate on social impact and the use of social impact methods. Maas & Liket (2011) found that different terms used by different social searchers is confusing. Words like “impact”, “output”, “effect”, “outcome” and “social return” like to be exchangeable (Maas & Liket, 2011). The term (social) impact is often replaced by terms such as “social value creation” (Emerson et al. 2000) and “social return” (Clark et al., 2004).

Therefore, social impact can also be defined as: “social value that is created when resources, inputs, processes or policies are combined to generate improvements in the lives of individuals or society as a whole” (Emerson et al. 2000). By impact we mean the proportion of the total outcome that happened as a result of the activity of the venture, above and beyond what would have happened anyway (Clark et al., 2004).

In recent years there is more consensus on the definition of social impact: The reflection of social outcomes as measurements, both long-term and short-term, adjusted for the effects achieved by others (alternative attribution), for effects that would have happened anyway (deadweight), for negative consequences (displacement), and for effects declining over time (drop-off) (Clifford, 2014). Social impact is defined in reference to four key elements (Clifford, 2014) as:

1. the value created as a consequence of someone’s activity (Emerson et al., 2000); 2. the value experienced by beneficiaries and all others affected (Kolodinsky et al., 2006); 3. an impact that includes both positive and negative effects (Blankenburg, 1995);

4. an impact that is judged against a benchmark of what the situation would have been without the proposed activity. (Clark et al. 2004).

M

From the field of Social Psychology, social impact is related to the concepts: strength, immediacy and number of sources (Latane, 1981). Strength, immediacy and number of sources are important but how to link that to the social media activity? Can we link likes directly to positive attitudes? Recent

evidence suggests that having many “likes” does not necessarily translate into more positive brand attitudes (John et al. 2016) or purchases (Lake, 2011). In addition, social impact theory (Latane, 1981) advocates that having a large number of supporters (e.g., fans, followers) does not imply more

positive brand attitudes and higher purchase intent. Therefore, it is important to look to other methods to measure social impact.

Maas & Liket (2011) and Grieco et al. (2015) both use the impact value chain of Clark et al. (2004) as the best method to measure social impact. This definition is based on the so-called Impact Value Chain and is developed to differentiate outputs from outcomes and impact. The impact value chain distinguishes outputs, outcomes and impacts as measures of effectiveness. Efficiency, the other component of performance, is defined as ratio of outputs to inputs. In the proposed framework, the levels of integrated management - operative, strategic and normative - are logically combined with the performance levels output, outcome and impact (Achleitner et al. 2009).

Impact value chain

In the figure below the impact value chain is visualized. A brand in social media is investing an amount of input to do social media activities which result in outputs as posting a brand post or responding on someone’s comments. The outputs of organizational social media activities can be measured by counting the number of posts and comments (or other activities). Outcomes can be specific changes in attitudes, behaviors, knowledge, skills, status, or level of functioning that result from the activities, such as care for someone, avoiding bad situations, finding solutions for a specific situation (Clark et al. 2004). Related to social media, liking the post, give a reaction, or share the post is one of these

outcomes. Hopefully, the outcomes are aligned with the goals. Then, the changed situation is returned

to the situation until the organization is using their input again and the process is following the same path.

Impact value chain (Clark et al. 2004) Because social impact can be positive or negative, the sentiment of a social media post is not

important for the size of social impact, but can be important for the social goal. In this study the effect of negative or positive contributions to social impact is ignored. According to the definition there is a need to measure the value created (eg. shares / comments) and the value experienced (eg. likes). Before we can really measure the impact, we need to measure the social media outcomes. In the next sections of this chapter we will describe what the best methods is to measure social media outcomes.

S

Lot of research is done how to measure outcomes or performance of social media activities for brands, especially in the concepts of Social Media ROI, Customer Engagement and Word of Mouth (WOM). In the next sections the Social Media ROI and Customer Engagement concepts are reviewed and compared with the broadly used Facebooks engagement score method to measure social media outcomes.

FACEBOOK ENGAGEMENT RATE

The Facebook engagement rate method: interactions per mille (IPM+) is an easy way to measure engagement. The Facebook engagement rate is a broadly used tool to measure engagement in practice.

The measure sums the interaction of likes, comments, shares, clicks and reach and divides that number by the number of page fans. In the past the score was known as the IPM measure, when the shares were not part of the measure. Facebook updated the tool and renamed the measure to IPM+ when adding the shares variable to the measure.

In this research we are searching for a measure for total impact, thus not impact per page fan or per mille. In the Facebook engagement rate measure is a share is as important as a like. However, there is some kind of weighting in the numbers because someone who shares, is also part of the reach, likely part of the likes, and part of the shares. Therefore, a share is likely counted two or three times in the Facebook IPM measure.

The Facebook IPM measure is probably a too simple tool (because there is no further weighting of the different kind of interaction types, and there are probably missing variables) to really measure the social media outcomes, we explore the concepts of Social media ROI and Customer Engagement to measure social media outcomes.

SOCIAL MEDIA ROI

Social Media ROI is another tool to measure social media outcomes. Hoffman & Foder (2010) developed a method to measure social media ROI which combines metrics to measure brand awareness, engagement and word of mouth in social media.

Social media efforts that are developed in the context of the 4c’s — connections, consumption, creation and control — that underlie consumer motivations to participate will lead to higher ROI because the company’s marketing investments can better leverage the active “investments” its customers will make as they engage with the company’s brands.

Four perspectives are needed to achieve the highest ROI. Consumers, followers and friends “consume” or view the content, comment on it (“create”), follow (“connect”) and share it with their friends and anyone else and provide their uncensored thoughts about it (“control”) for any and all to view. In the table below, Hoffman & Foder (2010) give variables which can be used to measure brand awareness, brand engagement and word of mouth. While it is not exhaustive, it gives a good guideline how to measure brand awareness, brand engagement and word of mouth.

Table of variables to measure brand awareness, brand engagement and word of mouth (Hoffman & Foder, 2010) A simple formula that guides on how to publish and what to measure in social media is not possible due to the particular circumstances of each brand and because of the very distinct set of goals and possibilities every business has (Sabate et al., 2014).

According to the method of Hoffman & Foder (2010), as a first step, marketers should focus on objectives that explicitly recognize the value of operating in the social media environment. Is the brand active on social media to achieve brand awareness or brand engagement or word of mouth? Based on the operating goals, the weights of different social media measures can also be different.

Brand awareness

Brand awareness refers to the strength of a brand’s presence in consumers’ minds (Pappu et al. 2007). Rossiter and Percy (1987) defined brand awareness as the consumers’ ability to identify or

recognize the brand. If the brand is more aware in the minds of customers or followers, then there is a better connection between the brand and the customer / follower. Brand awareness is linked to the first C: connection. Measurement of brand awareness for a Facebook brand page can be done by the number of fans (# followers) and the number of impressions per post (# impressions).

Brand engagement

Brand engagement refers to the level of an individual customer's motivational, brand-related and context-dependent state of mind characterized by specific levels of cognitive, emotional and behavioral activity in direct brand interactions (Hollebeek, 2011). Brand engagement is the level of connection which results in creation, consumption and control between the customer and the brand. Brand engagement is linked to 3 C’s: consumption, creation and control. Measurement of brand engagement for a Facebook brand page post can be done by the number of comments, active users, interaction and the impressions-to-interaction ratio.

Word of Mouth

Word of Mouth (offline) or electronic e-WOM (online) refers to any positive or negative statement made by potential, actual, or former user about a subject, product or company, which is made available to a many/thousands of people and institutions (Hennig-Thurau, 2004). There is a

differentiation between firm-generated content and user-generated content (Goh et al. 2012). When social media users comment / share on firm-generated content, then they do an active Word of Mouth (WOM) investment in the company brand. Once consumers are aware and engaged, they are in a position to communicate their opinions to other consumers. Satisfied and loyal consumers

communicate their positive attitudes toward the brand itself or toward the social application created by the company to new, prospective customers both online and offline. In that situation, this is resulting in a higher ROI for the company (Hoffman & Foder, 2010).

The concept of word of mouth is seen as cheaper and more effective online with social media than by traditional media, but their utility hinges on people transmitting content that helps the brand. If no one shares a company’s content, or if consumers share content that portrays the company negatively, the benefit of social transmission is lost (Berger & Milkman, 2012). We are more likely to trust our friends. And when we share, we select people who we think would find that given piece of information most relevant. So, word of mouth tends to reach people who are actually interested in the topic being discussed (Berger, 2013).

Word of Mouth related to the fourth C, control. Customers / followers take control of the message and share the statement. The level of Word of Mouth is related to the number of impressions /

appearances in timeline of friends. And the number of posts on the wall. Word of Mouth for a

Facebook brand page post can be measured by the number of shares, the impressions of the shares, and the number of posts on the wall. Although we only measure eWOM with these measures are not WOM overall, because over 80% of word of mouth is offline face-to-face (Keller & Fay, 2012). Although Hoffmann & Fodor (2010) gives not an exhaustive model, it gives guidance about which measures a brand manager can use to measure social media ROI. Hoffman & Foder (2010) clearly mention as first step to select a specific goal, to improve brand awareness, brand engagement and/or word of mouth. Based on the goal, the weights of different measures can also be different. We need an applied method to measure the social media outcomes. The method of Social Media ROI cannot be applied in general to measure social media outcomes, it will need adjustments every time because of the goal adjustment.

The guideline of Hoffmann & Fodor (2010) gives insight in more variables to measure social media outcomes. Much more variables can be measured than which are used in the Facebook IPM+ measure. In this research we focus on measuring the impact of a specific FGC post for a brand. A very good measure to find that out is probably the new followers after a brand post, this variable is missing in the Facebook measure. But Hoffmann & Fodor (2010) is not giving a clear method to measure social outcomes in general. Therefore, we will review the next measure: Customer Engagement (CE).

CONSUMER ENGAGEMENT

Customer Engagement (CE) is another concept to measure social media outcomes. As companies try to engage customers better, researchers have attempted to understand CE empirically. CE goes

beyond purchase and is the level of the customer’s (or potential customer’s) interactions and

connections with the brand or firm’s offerings or activities, often involving others in the social network created around the brand/offering/activity (Vivek et al., 2014). Even scales are developed for

measuring CE.

Customer engagement (CE) is defined by marketing literature as “the mechanics of a customer’s value addition to the firm, either through direct or/and indirect contributions, where direct contributions consist of customer purchases, and the indirect contributions consist of incentivized referrals that the customer provides, the social media conversations that customers have about the brand and the customer feedback/suggestions given to the firm” (Pansari & Kumar, 2017). Multiple studies have explored its nature, dimensions, and fundamental propositions (Brodie et al. 2011; Vivek et al. 2012), proposed frameworks (Hollebeek et al. 2016; Pansari & Kumar, 2016; Muntinga et al., 2011), and suggested its applicability in different platforms and fields of business (Kumar and Pansari 2016). CE is a source of competitive advantage. It is positive affecting the company’s performance. A

customer’s emotional connection with a company is key to achieving this advantage (Kumar & Pansari 2016). CE has been discussed in marketing literature (Vivek et al., 2012) as an outcome measure of the firm’s activities. If it possible to measure CE online, then the outcome of social media activities can be measured.

According to the method used by Simon & Tossan (2018) to measure consumer engagement on a brand's Facebook page, the consumers’ online brand-related activities (COBRA) typology from Muntinga et al. (2011) is used to combine several measures with weights. Muntinga et al. (2011) uses three dimensions labeled as follows: Consuming, Contributing, and Creating.

The first COBRA dimension with the least level of brand related activity is consuming, the second is contributing and as last dimension with the most brand related activity is creating.

Consuming covers all activities that are associated with a minimum level of online brand-related activeness. Consuming denotes relatively passive activities that involve participating without actively contributing or creating brand-related content.

The middle level of online brand-related activeness, contributing covers both consumer-to-content and consumer-to-consumer interactions about brands. Such moderately active behaviors include for instance participating in brand-related conversations on social networking sites, forwarding brand content, commenting on brand-related weblogs, and rating products/brands and other consumers’ brand related contributions.

Creating is the highest level of brand-related activeness on social media. Consumers who create brand-related content are actively producing and publishing the content that other consumers consume and contribute to. The table below shows metrics that are partly based on how Simon & Tossan, 2018 applied the COBRA typology to Facebook pages.

COBRA type Level Metric Description

Consuming 1 #impressions direct and indirect impressions of the

brand’s Facebook Post

#clicks clicks on the brand’s Facebook post

Contributing 2 #likes likes on this brand’s Facebook post

#engagedincomments engaged in conversations with community members on this brand’s Facebook post

#comments commented user comments on the

brand’s Facebook post

Creating 3 #shares shares the post off this brand's Facebook

page on own timeline

According to the COBRA typology a share is more brand related, than a comment, and a comment is more brand related than an impression or click. The mathematical formula for CE (based on Muntinga et al., 2011) is (where c1 < c2 < c3):

The weights c1, c2 and c3 are unknown. But according to the brand-related activeness in the COBRA framework we expect c3 > c2 > c1. These weights are tested in the hypothesis testing chapter of this thesis.

The COBRA typology can be used as method to measure social media outcomes. The theoretical framework to measure social impact is built on the Impact Value Chain of Clark et al. (2004). The COBRA typology with a customer engagement scale of Muntinga et al. (2011) is used to measure impact through social media activities. In social media there are activities (e.g. creating brand posts) that result in outcomes (e.g. interaction, comments, shares, etc) which are hopefully aligned with the goals which are set. Thereafterthis changed situation is returned to the input again and is following the same path. A measurement tool for social outcomes, results in a measurement tool of social impact as well. The model of Clark et al. 2004 consists of the following formula: social outcomes minus what should have happened anyway = social impact.

I

In this thesis three out of numerous concepts are considered to improve the social impact in social media: Top of mind, Extra information and Interactivity. These concepts have already known and tested results in literature. We use this information to test the theoretical framework to measure social impact.

T

-

A

Top of mind is a concept which mentioned in literature (Berger, Schwartz, 2011; Berger & Ivengar, 2013) to boost word of mouth. Word of mouth is closely related to customer engagement (Vivek, Betty and Morgan, 2012). In the literature review above, word of mouth is also discussed as part of social media objectives to measure ROI. However, there is no evidence that there is higher social impact with top of mind posts.

If people have less time to think about what to say in oral communication, they should be more likely to talk about things that are top of mind (Berger & Iyengar, 2013). For example, the Olympic Games, if something is happening there where you can post about, then it is likely to have more effect on the moment the Olympic Games are on, than a year later, when the topic is not top of mind anymore. About actuality it is generally known that it is working great on social media. News media are

increasingly using Facebook as a platform for distribution and user interaction (Hille & Bakker, 2013). Is there a relation between top of mind posts and social impact?

We expect that a higher top of mind score is influencing social impact positively.

E

A link to a website is more interactive (Fortin and Dholakia, 2005) since brand fans can click on that link. The search for information explains why people engage in brand-related activities on social media (Muntinga et al., 2011). Extra information should lead to higher social impact.

We expect that extra information is influencing social impact relatively positively.

I

Another way to improve the social impact of a brand post is interactivity. Interactivity is characterized by two-way communication between companies and customers, as well as between customers themselves (Hoffman & Novak, 1996). Interactivity is part of the customer engagement measures. Brand post characteristics differ in the degree of interactivity; a brand post with only text is not at all interactive, while a link to a website is more interactive since brand fans can click on that link (Fortin & Dholakia, 2005). Since the objective of brand posts is to motivate brand fans to react (i.e., liking and/ or commenting), we expect that higher degrees of interactivity will generate more likes and comments. A question is logically asking for an answer. A question in a post acts as a highly interactive brand post characteristic because it begs an answer from brand fans. Research shows inconclusive findings (negative or no effect versus positive effect) regarding interactivity on outcome measures (de Vries et al. 2012).

We expect that a slightly higher interactivity score (thus by asking feedback) which influences social impact positively, but not significant.

CONCEPTUAL FRAMEWORK AND HYPOTHESES

The theoretical framework leads us to a way to measure impact of social media posts. In the proposed framework we want to test the effects of adding extra interactivity by asking a question, testing what the result is of top of mind messages, and if extra information with a hyperlink is giving a positive result to a social brand post impact. The graphical representation of this research is presented below:

Weights

Consuming, contributing and creating are weighted to measure social impact. Five possible weights sets are c=[1,1,1], c=[1, 0.25, 0.1], c=[0.1, 0.25, 1] ], c=[0.1, 5, 10], c=[0.01, 5, 10]. These weight sets for the c’ variables are tested in the hypothesis testing section.

We aim to measure a difference in significance against the standard Facebook engagement rate and impact levels which are manually labeled. Averages are compared in the testing which of the weight sets is the best performing. Based on previous literature results, we expect that the first concept ‘Top of mind’ has a strong expected effect on social impact, the second concept ‘extra information’ has an average expected effect on social impact, and the third concept ‘interactivity’ has not a significant expected effect on social impact.

R

Top of mind

Actuality and top of mind posts are likely to work better because if people have less time to think about what to say in oral communication, they should be more likely to talk about things that are top of mind (Berger & Iyengar, 2013). We also expect this behavior online. For example, the Olympic Games, if something is happening there where you can post about, then it is likely to have more effect on the moment the Olympic Games are on, than a year later, when the topic is not top of mind anymore.

H1a: The social impact of a social media post with a top of mind topic in a post is higher. H1b: The social impact of a social media post without a top of mind topic in a post is lower.

Extra information

A link to a website is more interactive (Fortin and Dholakia, 2005) since brand fans can click on that link. But does it overall also create more social impact? I propose that the overall impact of a post with a hyperlink is higher.

H2a: The social impact of a social media post with a hyperlink to a website is higher. H2b: The social impact of a social media post without a hyperlink to a website is lower.

Interactivity

Questions are logically asking for an answer and contribution. Therefore, we expect a higher Social Impact. Research shows inconclusive findings (negative or no effect versus positive effect) regarding interactivity on outcome measures (de Vries et al. 2012). Although their finding is that by asking a question in a post, the number of likes will be lower, other measures will be higher (comments and shares) our hypothesis is that it could rise the impact effect.

H3a: The social impact of a social media post with a question in the title of the hyperlink is higher. H3b: The social impact of a social media post with a question in the title of the hyperlink is lower. H3c: The social impact of a social media post with a question at the end of a post is higher. H3d: The social impact of a social media post with a question at the end of a post is lower.

ORGANIZATION

This thesis is using a specific brand Ik Mis Je (IMJ) to evaluate the conceptual framework. The primary goal of IMJ is to help people share their experience with mourning about their loved one. IMJ is using several cross-media to achieve this primary goal. Television, Radio and Internet, more specifically, a social media community on Facebook is used to reach, connect and share the experience of this group of people. IMJ is using next to Facebook page community a website to post blogs, extra video’s and ability to post a personal story. Most social media posts on Facebook of IMJ are linked through to the website of IMJ.

A quote of a visitor: “I wanted that IMJ was there 9 years ago, when I lost my husband. In that time, I had to find out everything by myself. If your community was there at that time I had found much more possibilities to mourn, and was it more acceptable for me what I felt and did.” Impact and reach are relevant concepts for IMJ, because impact and reach are the basis of success for IMJ.

Because the social impact is very well linked to the goal of IMJ, this brand is a good fit for testing the conceptual framework.

Evangelische Omroep (EO) is a media organization which is visible in society through television, radio, internet, magazines and events. EO is part of the Dutch Public Broadcasting system. It distinguishes itself clearly from all other associations within the system through a unique mission. The employees have the desire to shape and pass on the good news of God's Kingdom from a personal relationship with Jesus Christ. With this desire, the EO wants to reach as many people as possible and build on society. IMJ is a brand of EO.

HYPOTHESIS TESTING

In this chapter, we focus on the hypothesis testing of the conceptual model with the data gathered by IMJ. We aim to find out that our new impact score is giving the same or equally likely significance as the impact score (manually labeled by the editors of IMJ) for the top of mind, extra information and interaction concepts as found in literature. We aim to find out that the significance of one of our impact scores is more equal to the significance of the manual labeling of impact than the reference Facebook IPM score.

A number of 730 posts from June 2017 till August 2018 is used to test the hypothesis. Only posts without sponsoring are considered to be tested in this research. Two ordinal scores were asked from the participants. The participants were two social media editors of IMJ.

An ordinal impact score (0-10) and a ordinal top-of-mind score between (0-10) were asked to evaluate the conceptual model and hypothesis. All posts with null values are ignored. All posts that had

sponsoring in Facebook were filtered out.

ANOVA is used to evaluate the significance levels of the facebook IPM, conceptual framework and manually labeled impact levels. We carry out one sided t-tests on all of our hypothesis. Spearman Rank Correlation, Pearson R and Kendall’s tau is used to find the correlation of the calculated impact scores with the labeled impact score.

R

First, we find out which of the calculated impact scores is having the best correlation to the manually labeled impact scores. This is done by using the Spearman Rank Correlation coefficient.

Correlation coefficients

Correlation coefficients between .10 and .29 represent a small association, coefficients between .30 and .49 represent a medium association, and coefficients of .50 and above represent a large association or correlation. The fifth weight set (c=[0.01,5,10]) is performing the best.

Hypothesis testing

H1a: The social impact of a social media post with a top of mind topic in a post is higher.

The higher the top-of-mind score the higher the average of the impact scores (third column). Facebook IPM is the reference interaction score of Facebook without the divide by number of followers part of the normal Facebook IPM+ score. And the last five columns are the Impact scores calculated by the formula of the conceptual model with different weights.

Based on the input of the editors of IMJ, we considered a top of mind score >= 6 as higher than normal.

Null hypothesis: Impact = 6.21 (with a top of mind score >= 6) Alternative hypothesis: Impact > 6.21 (with a top of mind score >= 6)

We conclude that there is enough statistical evidence that indicates that the average impact is greater than 6.21 with a top of mind score equal or greater than 6.

H1b: The social impact of a social media post without a top of mind topic in a post is lower.

Null hypothesis: Impact = 6.21 (with a top of mind score < 6) Alternative hypothesis: Impact < 6.21 (with a top of mind score < 6)

We conclude that there is enough statistical evidence that indicates that the average impact is lower than 6.21 without a top of mind score greater or equal than 6.

H2a: The social impact of a social media post with a hyperlink to a website is higher.

Null hypothesis: Impact = 6.21 (with extra information = Y) Alternative hypothesis: Impact > 6.21 (with extra information = Y)

We conclude that there is not enough statistical evidence that indicates that the average impact is greater than 6.21 with extra information.

H2b: The social impact of a social media post without a hyperlink to a website is lower.

Null hypothesis: Impact = 6.21 (with extra information = N) Alternative hypothesis: Impact < 6.21 (with extra information = N)

We conclude that there is not enough statistical evidence that indicates that the average impact is lower than 6.21 without extra information.

H3a: The social impact of a social media post with a question in the title of the hyperlink is higher.

Null hypothesis: Impact = 6.21 (with question in post title) Alternative hypothesis: Impact < 6.21 (with question in post title)

We conclude that there is not enough statistical evidence that indicates that the average impact is higher than 6.21 with a question in the post title.

H3b: The social impact of a social media post with no question in the title of the hyperlink is lower.

Null hypothesis: Impact = 6.21 (with no question in post title) Alternative hypothesis: Impact < 6.21 (with no question in post title)

We conclude that there is not enough statistical evidence that indicates that the average impact is lower than 6.21 with no question in the post title.

H3c: The social impact of a social media post with a question in the message is higher.

Null hypothesis: Impact = 6.21 (with question in post) Alternative hypothesis: Impact < 6.21 (with question in post)

We conclude that there is not enough statistical evidence that indicates that the average impact is higher than 6.21 with a question in the post.

H3d: The social impact of a social media post with no question in the message is lower.

Null hypothesis: Impact = 6.21 (with no question in post) Alternative hypothesis: Impact < 6.21 (with no question in post)

We conclude that there is not enough statistical evidence that indicates that the average impact is lower than 6.21 with no question in the post.

DISCUSSION

Our newly proposed 6-variable impact scale appears to have a good validity and is better than the Facebook IPM score to measure social impact. According to the Spearman rank correlation coefficient the impact score with the fifth weight set (c=[0.01, 5, 10]) performed better than the others, and all the impact scores performed better than the Facebook IPM score. According to the hypothesis testing the third weight set is giving the most equal significance results to the hypothesis which were tested. The current way of measuring impact in organizations is currently mainly based on the standard facebook IPM method. This method has a lack of weighting the different interaction aspects in social media. This study provided insights that weighting is could be important to get closer to the real impact of a social media post. However, it is difficult to measure something without a ground truth. This study is based on input from IMJ.

Consistent with other studies, this study shows that a top-of-mind post is significantly improving the impact of the post. Therefore, it is important for organization to plan, and check if a post is posted on the right moment in time of the day, week or year. This study did not find a solid significant relation between extra information and impact, where other studies did find an average relation with

engagement. Also consistent with other studies is the use of interactivity in social media posts. Asking for interactivity is not significantly raising engagement or impact.

This study is using the concept of consumer engagement as basis for measuring social outcomes and therefore social impact. A logical conclusion could be that there is an average relation between engagement and social impact.

Although we did not fully test the weights and formula of the conceptual model, the hypothesis which were evaluated in the previous section resulted in the same conclusions as the impact score which was given by the editors of IMJ. Further research should be done on the weights and if all the parameters of the formula are needed or to include more variables to calculate the social impact of a social media post.

LIMITATIONS & FURTHER RESEARCH

As is the case with any research, readers need to consider the presented results within the context of limitations. Also, the process of posing and answering particular research questions typically

generates more questions that need to be explored through further research.

The definition, formula and weights of social impact is derived from literature, but the derivation is not tested with a big data set and with multiple Facebook brand pages. This research uses a convenience sample, which is adequate in size to provide validity for its findings for one brand, but limits their generalizability beyond the present study. A bigger dataset from different Facebook pages is needed to evaluate and find the best weights for the c’ weights in the conceptual model.

The different types of impact, like positive and negative social impact (social media sentiment) is not studied in this thesis.

Social ROI gave an insight to also use brand awareness (how many people are connected to the Facebook page) into the impact scores. In the current research this number is not considered as variable, because it was not available.

Benchmarking with other brands could be a method to compare and evaluate the social media impact results of a brand with another brand and to set goals to accomplish.

Finally, more research needs to be carried out to get to a measurable impact score for social media. A particular focus on the weights in multiple Facebook pages appears to be desirable.

ACKNOWLEDGEMENT

I would like to thank my thesis advisor Gijs Overgoor of the Amsterdam Business school at University of Amsterdam. Gijs responded on my emails within a split second, and he provided me with ideas throughout the whole process. This was a very pleasant experience. He motivated me in a good way to work on my thesis when I was mourning about my brother.

I would also like to thank Albert-Jan Schol and Heleen Dekens of IMJ who were involved in the setup of this thesis idea and helped me with providing the data. Without their passionate participation and input, this thesis could not have been successfully worked out.

Finally, I express my very profound gratitude to my wife for providing me with unfailing support and continuous encouragement throughout my two years of study and through the process of researching and writing this thesis. This accomplishment would not have been possible without her. Thank you.

REFERENCES

Achleitner, A. & Bassen, A., Roder, B. & Spiess-Knafl, W. (2009). Reporting in Social Entrepreneurship. SSRN Electronic Journal. 10.2139/ssrn.1493266.

Bagnoli, L., & Megali, C. (2011). Measuring performance in social enterprises. Nonprofit and Voluntary Sector Quarterly, 40, 149-165.

Berger, J., Schwartz E.M. (2011). What Drives Immediate and Ongoing Word of Mouth. Journal of Marketing Research, Vol. 48, No. 5 (October 2011), pp. 869-880

Berger, J., Milkman K.L. (2012) What Makes Online Content Viral? Journal of Marketing Research: April 2012, Vol. 49, No. 2, pp. 192-205.

Berger, J., Iyengar, R. (2013). Communication Channels and Word of Mouth: How the Medium Shapes the Message, Journal of Consumer Research, Volume 40, Issue 3, 1 October 2013, Pages 567–579, https://doi.org/10.1086/671345

Berger, J. (2014). The Fascinating Psychology behind Word-of-Mouth Marketing, SXSW, Article by Nicole Carter

Blankenburg, F. (1995). Methods of Impact Assessment Research Programme: Resource pack and discussion. The Hague: Oxfam UK/I and Novib

Brodie, R. J., Hollebeek, L. D., Juric, B., & Ilic, A. (2011). Customer engagement: conceptual domain, fundamental propositions, and implications for research. Journal of Service Research, 14(3) 252-271. Booth N., Matic J.A., (2011) "Mapping and leveraging influencers in social media to shape corporate brand perceptions", Corporate Communications: An International Journal, Vol. 16 Issue: 3, pp.184-191, https://doi.org/10.1108/13563281111156853

Clark, C., Rosenzweig, W., Long, D., Olsen, S. (2004). Double bottom line project report: Assessing social impact in double bottom line ventures. Methods catalog,

http://www.riseproject.org/DBL_Methods_Catalog.pdf

Clifford, J. (2014). Impact Evaluation by Social Enterprises: Measuring the un-measurable?, 10th annual meeting of the OECD LEED forum on partnerships and local development, Stockholm, April 2014

Ebrahim, A. (2010). The many faces of nonprofit accountability. The Jossey-Bass handbook of nonprofit leadership and management (pp. 101-123).

Fortin D.R., Dholakia R.R. (2005). Interactivity and Vividness Effects on Social Presence and

Involvement with a Web-Based Advertisement. Journal of Business Research, 58 (3) (2005), pp. 387-396

Goh, K.Y., Heng, C.S. and Lin, Z. (2013). Social media brand community and consumer behavior: quantifying the relative impact of user-and marketer-generated content, Information Systems Research, Vol. 24, No. 1, pp.88–107.

Grieco C., Michelini L., Lasevoli G. (2015). Measuring value creation in social enterprises a cluster analysis of social impact assessment models. Nonprofit and voluntary sector quarterly 44 (6), 1173-1193

Gruzd, A., Wellman, B. Sc Takhteyev, Y. (2011). Imagining Twitter as an imagined community. American Behavioral Scientist, 55, 10, pp. 1294-1318.

Hennig-Thurau T., Gwinner, K. P., Walsh, G., Gremler D.D.(2004) Electronic word-of-mouth via consumer-opinion platforms: What motivates consumers to articulate themselves on the Internet?. Journal of Interactive Marketing, Volume 18, Issue 1, Pages 38-52, ISSN 1094-9968,

https://doi.org/10.1002/dir.10073.

Hille, S., Bakker, P. (2013). I like news. Searching for the ‘Holy Grail’ of social media: The use of Facebook by Dutch news media and their audiences. European Journal of Communication. Vol 28, Issue 6, pp. 663 - 680

Hoffman, D. L., & Novak, T. P. (1996). Marketing in Hypermedia Computer-Mediated Environments: Conceptual Foundations. Journal of Marketing, 60(3), 50

Hoffman, D & Fodor, M. (2010). Can You Measure the ROI of Your Social Media Marketing?. MIT Sloan Management Review. 52.

Hollebeek, Linda D. (2011) Demystifying customer brand engagement: Exploring the loyalty nexus, Journal of Marketing Management, 27:7-8, 785-807, DOI: 10.1080/0267257X.2010.500132

ICGP. (2003). Guidelines and Principles for Social Impact Assessment in the USA. Interorganizational Committee on Principles and Guidelines for Social Impact Assessment

Jahn B., Kunz W. (2012) "How to transform consumers into fans of your brand", Journal of Service Management, Vol. 23 Issue: 3, pp.344-361, https://doi.org/10.1108/09564231211248444

John L. K., Emrich O., Gupta S., Norton M.I. (2017) Does “Liking” Lead to Loving? The Impact of Joining a Brand’s Social Network on Marketing Outcomes. Journal of Marketing Research: February 2017, Vol. 54, No. 1, pp. 144-155.

Kaplan, A. M., & Haenlein, M. (2010). Users of the world, unite! The challenges and opportunities of Social Media. Business Horizons, 53, 59–68. doi:10.1016/j.bushor.2009.09.003

Keller, E., & Fay, B. (2012). Word-of-mouth advocacy: A new key to advertising effectiveness. Journal of Advertising Research,52, 459–464. https://doi.org/10.2501/JAR-52-4-459-464

Kimmel, Allan J., & Kitchen, Philip J. (2014) Word of mouth and social media, Journal of Marketing Communications, 20:1-2, 2-4, DOI: 10.1080/13527266.2013.865868

Kolodinsky, J., Stewart, C., Bullard, A. (2006). Measuring economic and social impacts of membership in a community development financial institution, Journal of Family and Economic Issues, 27(1): 27-47 Kumar, A., Bezawada, R., Rishika, R., Janakiraman, R & Kannan, P.K. (2015). From Social to Sale: The Effects of Firm Generated Content in Social Media on Customer Behavior. Journal of Marketing. 80. 10.1509/jm.14.0249.

Kumar, V., Pansari A. (2016), “Competitive Advantage Through Engagement.” Journal of Marketing Research,53 (4), pp. 497-514.

Lake, A. (2011), “Why Facebook Fans Are Useless,” iMedia Connection

Laroche M., Habibi M.R., Richard M. (2013) To be or not to be in social media: How brand loyalty is affected by social media?, International Journal of Information Management, Volume 33, Issue 1, ISSN 0268-4012, https://doi.org/10.1016/j.ijinfomgt.2012.07.003.

Latané, B. (1981). The psychology of social impact. American Psychologist, 36, 343–365

Lovejoy, K. & Saxton, G. D. (2012), Information, Community, and Action: How Nonprofit Organizations Use Social Media. Journal of Computer-Mediated Communication, 17: 337-353. doi:10.1111/j.1083-6101.2012.01576.x

Lu, Y, Zhao, L. Sc Wang, B. (2010). From virtual community members to c2c e-commerce buyers: trust in virtual communities and its effect on consumers’ purchase intention. Electronic Commerce Research & Applications, 9, 4, pp. 346-360.

Maas, K., & Liket, K. (2011). Social impact measurement: Classification of methods. In R. Burritt, S. Schaltegger, M. Bennett, T. Pohjola, & M. Csutora (Eds.), Environmental man- agement accounting and supply chain management, eco-efficiency in industry and science (Vol. 27, pp. 171-202) Delft, The Netherlands: Springer.

Muntinga D.G., Moorman M. & Smit E.G. (2011). Introducing COBRAs, International Journal of Advertising, 30:1, 13-46, DOI: 10.2501/IJA-30-1-013-046

Nolan, L. (2015). The impact of executive personal branding on non-profit perception and communications. Public Relations Review, 41, 288–292. doi:10.1016/j.pubrev.2014.11.001

Nowak, A., Szamrej, J., Latane, B. (1990). From Private Attitude to Public Opinion: A Dynamic Theory of Social Impact. Psychological Review. 97(3):362-376

Oh, H.J., Ozkaya, E. & Larose, R. (2014). How Does Online Social Networking Enhance Life Satisfaction? The Relationships Among Online Supportive Interaction, Affect, Perceived Social Support, Sense of Community, and Life Satisfaction. Computers in Human Behavior. 30. 69–78. 10.1016/j.chb.2013.07.053.

Pappu, R., Quester, P.G. & Cooksey, R.W. J Int Bus Stud (2007). Consumer-based brand equity: improving the measurement – empirical evidence. https://doi.org/10.1057/palgrave.jibs.8400293 Park N., Kee K.F., Valenzuela S. (2009) Being Immersed in Social Networking Environment:

Facebook Groups, Uses and Gratifications, and Social Outcomes, CyberPsychology & Behavior 2009 12:6, 729-733

Preece, J. & Maloney-Krichmar, D. (2005). Online Communities: Design, Theory, and Practice. J. Computer-Mediated Communication. 10. 10.1111/j.1083-6101.2005.tb00264.x.

Rossiter, J.R. and Percy, L. (1987). Advertising and Promotion Management, McGraw-Hill, New York, NY.

Sabate, F., Berbegal-Mirabent, J., Cañabate, A., Lebherz, P.R. (2014). Factors influencing popularity of branded content in Facebook fan pages. European Management Journal 2014, 32, 1001–1011. Simon F., Tossan V. (2018). Does brand-consumer social sharing matter? A relational framework of customer engagement to brand-hosted social media. Journal of Business Research, Volume 85, Pages 175-184, ISSN 0148-2963

Srinivasan, S., Rutz, O.J. & Pauwels, K. J. (2016). Paths to and off Purchase: Quantifying the Impact of Traditional Marketing and online Consumer Activity, Journal of the Academy of Marketing Science, 44 (4), 440–53.

Stephen A., Galak J. (2012). The Effects of Traditional and Social Earned Media on Sales: A Study of a Microlending Marketplace. Journal Of Marketing Research (JMR). October 2012;49(5):624-639 Vivek, S. D., Beatty, S. E., & Morgan, R. M. (2012). Customer engagement: Exploring customer relationships beyond purchase. Journal of Marketing Theory and Practice, 20(2), 122-146.

Vries de, L., Gensler, S., & Leeflang, P. S. (2012). Popularity of brand posts on brand fan pages: an investigation of the effects of social media marketing. Journal of Interactive Marketing, 26(2), 83-91. Wainwright S. (2002), Measuring impact: A guide to resources, London, NCVO Publications

Webster’s Ninth New Collegiate Dictionary. (1986). Merriam-Webster, Springfield, Mass.

Zanna, M. P., & Rempel, J. K. (1988). Attitudes: A new look at an old concept. in d. Bartal & A. W. Kruglanski (eds.), The social psy- chology of knowledge, 315–334. Cambridge, uK: Cambridge university Press.