Scrolling through your Facebook feed, you will likely encounter a barrage of advertisements, each vying for your attention and urging you to click on them. Like any other type of post on the platform, these advertisements permit users to express their reactions. It is expected to find that some of your friends have already interacted with these ads by ‘liking’ them.

Advertisers are keen on accumulating a significant number of such endorsements, particularly from individuals who are familiar to you, in the hope that this will increase the likelihood of you engaging with their content. However, intriguing new findings from Texas McCombs suggest that the effectiveness of this strategy can vary significantly depending on the nature of the advertisement and the relationship you have with the friend who has endorsed it.

The study, led by Ashish Agarwal, an associate professor of Information, Risk, and Operations Management (IROM), along with Andrew Whinston, a professor of IROM who also holds the Hugh Roy Cullen Centennial Chair in Business Administration and serves as the director of the Center for Research in Electronic Commerce at The University of Texas at Austin, has unveiled some nuanced insights into consumer behaviour on social media. Alongside Shun-Yang Lee from Northeastern University, the researchers concentrated on call-to-action (CTA) advertisements, which employ assertive language to prompt immediate action from users, such as purchasing or downloading a mobile app. This type of ad contrasts sharply with informational advertisements, which use more passive wording to invite users to ‘learn more’ about a topic or product.

While advertisers generally prefer CTA ads because they transition social media users directly into a buying mindset, previous studies have indicated that these ads can sometimes alienate users, particularly when they feel manipulated by the aggressive tone of the messaging. The researchers explored whether the accumulation of ‘likes’ could mitigate this aversion. Agarwal remarked on the dual nature of these assertive ads and the role social endorsements might play, questioning whether they would amplify or dampen user resistance.

To address these questions, the research team conducted a series of experiments. They initiated a field experiment by collaborating with a mobile app developer to place a CTA advertisement on Facebook, encouraging users to download an app. The ad was displayed over 710,000 times and garnered 799 ‘likes’ and 4,052 clicks. Additionally, a laboratory test assessed various combinations of advertisements and endorsements — comparing informational versus CTA ads and generic ‘likes’ versus ‘likes’ from friends — with each of the 982 participants listing five friends.

The results of these studies were telling. For informational ads, an increase in ‘likes’ correlated with a higher number of clicks, with the likelihood of clicking rising by 3% for every 100 generic likes and, even more significantly — 21% — for each ‘like’ from a friend. Conversely, the overall number of ‘likes’ for CTA ads did not considerably influence click rates.

However, ‘likes’ from friends did exert an influence, which varied depending on whether the user believed the friend had similar or dissimilar interests. ‘Likes’ from friends with similar interests increased the likelihood of a click by 180%, while those from friends with dissimilar interests reduced the odds by 66%. This discrepancy arose because users generally reacted negatively to the manipulative feel of CTA ads. However, when they perceived that alike had come from a friend with similar tastes, this endorsement lent greater credibility to the ad.

In a subsequent laboratory study, the team discovered that users were less antagonistic towards informational ads compared to CTAs. They found them less intrusive and thus more open to persuasion by ‘likes’.

These findings have important implications for advertisers and social media platforms that derive revenue from advertising. While displaying ‘likes’ might enhance engagement with informational ads, the same strategy could backfire with CTA ads. Agarwal suggests that both advertisers and social media platforms should consider making the presence of endorsements optional better to tailor their methods to the preferences of their audience, thereby enhancing the effectiveness of their advertising efforts.

More information: Ashish Agarwal et al, The Effect of Popularity Cues and Peer Endorsements on Assertive Social Media Ads, Information Systems Research. DOI: 10.1287/isre.2021.0606

Journal information: Information Systems Research Provided by University of Texas at Austin