What Drives Customer Engagement on Instagram?

March 1, 2021

Robert Rietveld, PhD Candidate, Willemijn van Dolen, PhD, Masoud Mazloom, PhD, and Marcel Worring, PhD

Visual social media platforms are rapidly growing in popularity; consumers collectively upload on average 42 million images and 1.65 billion likes each day on Instagram alone. Brands are a large part of this visual exchange as 80% of consumers follow at least one brand, and 70% of brands have an active Instagram account.1

Stock Photo of Hand Holding Cell Phone with Instagram App Open

Platforms like Instagram allow brands to create their own media and to engage with their followers by creating posts using elements of both visual and text. Advertisers can reach a large audience at relatively low cost compared to paid media activity. For example, Nike attracted 83 million followers over the last seven years (at the time of writing in 2019) and published 700 unique posts leading to different levels of likes and comments per post on social media sites.

Understanding what drives customer engagement across these posts is important for advertisers, since engagement initiatives have a proven positive impact on the bottom line. Our study explores how emotional and informative message appeals in visual and textual content influence customer engagement in terms of likes and comments.    

Emotional Appeals

Emotional appeals are defined in our research as visual and textual posts designed to make the consumer feel good about the brand or product and which lead to positive reactions and higher levels of recall compared to informative appeals. Emotions motivate and persuade consumers and often guide attitude formation and subsequent behavior. Images play an important role in this process as they capture the direct experience of objects they physically resemble and activate mental representations associated with strong emotional responses.

Emotional responses can be described in two main dimensions: arousal and valence (see Table 1 below). Arousal is the type of stimulus a viewer derives from the information encoded in the image. Valence refers to the positive or negative feelings incited by the information encoded in the image. Based on arousal and valence, there are four types of emotional appeals: Positive High Arousal, Negative High Arousal, Positive Low Arousal, and Negative Low Arousal.  

Chart Showing Relationships Between Positive and Negative Appeals

Consumers are especially influenced by emotions if they do not have any specific goal-directed basis for evaluation, as might be the case when engaging in hedonic browsing on social media. Because consumers process large numbers of posts in their social media streams, high-arousal visual appeals (compared to low-arousal visual appeals) can be a way to cut through the competitive social media clutter and get noticed. Visual emotional appeals were scored in this study by having subjects respond to a variety of adjective-noun pairs, such as “smiling children” or “young kids.”

Informative Appeals

Informative appeals focus on influencing consumers’ beliefs using arguments highlighted in the message context. For this reason, informative appeals tend to be fact-based and lean toward a more rational than emotional emphasis. Informative visual and textual appeals provide relevant information and reduce uncertainty; however, informative appeals can also entail an unwanted persuasion attempt or provide no new information to followers.

Visual informative appeals focus on the centrality of brand and product elements present in the visual content of the post. Visual product features displayed in images, for example, offer more concrete information than brand logos, as these logos can be attached to promotional activities or campaigns without displaying products. Visual brand elements facilitate recall of memories by attracting attention and activating positive or negative brand associations in memory. Textual informative appeals can be separated into eight groups: brand, product, deal, price, price comparison, segment, product availability, and product location. For example, a caption on Instagram may contain brand-generated text, including hashtags.


Customer response to visual social media is in large part contingent upon emotional content residing in the visual image as well as message content. Specifically, we distill four main findings from our results.

First, emotional appeals influence customer engagement more than informative appeals for both visual and text content. Second, the intensity level of arousal should be considered together with valence. Consider using positive high arousal appeals and negative low arousal to stimulate likes and comments.Findings suggest that effective facial expressions enjoy an unusual advantage compared to other visual stimuli. Third, apart from informative brand appeals, we find a negative influence of informative appeals on customer engagement. As consumers follow brands primarily for entertainment reasons, informative appeals, especially around the product, can signal an unexpected persuasion attempt and be considered manipulative. Fourth, an exception to the negative effect of informative appeals are visual brand centrality and textual brand mentions which positively contribute to comments and likes. Visual and textual brand elements may serve as a point of recognition for consumers which can activate positive brand associations consumers have established given prior brand experiences.

Managerial Insights

Managerial implications of this work for real estate agents and brokers include insights about how visual emotional and informative appeals might drive customer engagement. Our results provide some guidelines for what appeals are likely to impact likes and comments.

In advertising on social media, an agent or broker should focus on arousal and negative or positive psychological value assigned by viewers to another person, event, dream house, job, object, outcome, etc. in visual content. Positive high-arousal appeals such as ecstasy, amazement, admiration, and vigilance can enhance the contribution of consumers in terms of likes and comments. Negative low-arousal appeals, driven by the presence of faces, leads to higher customer engagement. Understanding how to elicit emotional arousal in marketing appeals is a prerequisite for successful real-estate customer engagement on Instagram.

Do not treat visual social media as primarily an informational channel. Informative appeals have less impact, and our results show that promoting products on Instagram does not increase customer engagement.  Informative appeals lead to a diminished number of likes and could increase annoyance and result in unfollowing of the agent or broker’s posts. Results highlight the opportunity to connect viewer emotions to the brand by emphasizing posts with brand mentions in the text.

Platforms like Instagram can reach large audiences at virtually zero cost, which has contributed to the popularity of social media among advertisers. Nevertheless, we recommend real estate professionals take advantage of more recent research findings to apply best practices in getting the desired results from social media.

. . . . . . . . . . . . . . . . . . .

Recommended Reading

Rietveld, Robert, Willemijn van Dolen, Masoud Mazloom, and Marcel Worring (2019), “What You Feel, Is What You Like Influence of Message Appeals on Customer Engagement on Instagram,” Journal of Interactive Marketing, 49, 20-53.

. . . . . . . . . . . . . . . . . . .



  1. Brain, Statistic (2015), “Instagram Company Statistics, Statistic Brain.” http://www.statisticbrain.com/instagram-company-statistics/.

. . . . . . . . . . . . . . . . . . .

About the Authors

Robert Rietveld, PhD Candidate
University of Amsterdam Business School
Robert Rietveld’s research examines how brand and user-generated content impact consumer decision making. He is an advocate of using the scientific method to answer business questions and realize growth. He was previously co-founder and managing director of a social media analytics agency that specialized in creating insights from digital consumer behavior.

Willemijn van Dolen, PhD
Professor of Marketing, University of Amsterdam Business School
Dr. Willemijn van Dolen (PhD – Maastricht University) focuses her research and teaching on online consumer communication, corporate social responsibility, and social media. She also has ample knowledge and experience in the area of big data, fintech, and behavioral economics. In addition to her current position, van Dolen is also a member of the Supervisory Board of Stichting Amsterdam Marketing, member of the NIMA (Netherlands Institute for Marketing) Scientific Advisory Council and member of the Supervisory Board of Starbucks Coffee EMEA B.V.

Masoud Mazloom, PhD
Post-Doctoral Researcher, Informatics Institute University of Amsterdam
Dr. Masoud Mazloom (PhD – University of Amsterdam) is a postdoc researcher at University of Amsterdam working with Prof. Marcel Worring and Prof. Willemijn van Dolen. Before starting his postdoc, he was PhD student at University of Amsterdam under supervision of Prof. Arnold Smeulders and Dr. Cees Snoek in the topic of multimedia video event detection by analyzing textual and visual data.

Marcel Worring, PhD
Director of the Informatics Institute and Full Professor of Business and Economics , University of Amsterdam Business School, Informatics Institute University of Amsterdam
Dr. Marcel Worring (PhD – University of Amsterdam) is a professor in data science for business analytics at the Amsterdam Business School and director of the Informatics Institute at UvA. During his PhD work he focused on image analysis both from a theoretical point of view to more applied shape analysis. From there he moved to document and video analysis. Currently, he is taking this a step further into multimedia analytics, multimedia analysis, multimedia mining, information visualization, and multimedia interaction.