Skip to main content

Using Social Media Engagement to Predict the Sale

Some social media engagements are more powerful than others.
Tim Halloran is a Senior Lecturer at Scheller College of Business, Georgia Tech.

Tim Halloran is a Senior Lecturer at Scheller College of Business, Georgia Tech.

Does liking lead to buying? Or does a positive comment mean a sale is more likely?

Marketers spend billions of dollars on social media in efforts to make connections with their customers—and make the sale. However, there is currently very little research regarding the type of Digital Consumer Engagement (DCE) associated with purchase frequency.  

“We saw an opportunity to fill that information gap,” said Scheller Senior Lecturer and Faculty Director of the EMBA program Tim Halloran. “We think our study offers valuable insights to businesses pinpointing which types of Facebook engagements are associated with incremental purchases—and which are virtually meaningless.”

Halloran and his co-author, Rich Lutz of the University of Florida shared their findings in a paper published in the Journal of Interactive Marketing. Titled "Let's Give Them Something to Talk About: Which Social Media Engagements Predict Purchase Frequency?", the paper investigates the relationship between Facebook engagements and incremental purchases, specifically identifying which engagements are more likely to predict a purchase.

Advancing a theoretical framework of social media engagement, the authors tested the association between various forms of DCE and customer behavior using archived field data. By matching store visit data from 1,066 loyalty program members of a leading national fast casual restaurant with their engagements on the brand's Facebook page, the authors compared the strength of the relationships between various forms of engagement and store visit frequency. Halloran and Lutz then identified the type of Facebook engagement that differentiated higher valued customers from average loyalty program members.


The study revealed that some Facebook engagements are significantly associated with increased or decreased store visits while others exhibit no discernible effects. Highlights from their published paper include:

  1. The strongest predictive power was found in Positive and Negative Comments. Both were significantly associated with post-engagement store visits (an increase for Positive, and a decrease for Negative).
  2. Liking a Facebook post had no significant impact on post-engagement visit frequency. Like, the most commonly used social media metric, is among the weakest forms of DCE, with no discernible relationship with purchase behavior. Indeed, they found it is not reliably different from no engagement at all.
  3. Somewhat surprisingly, taking the time to tag an individual with brand-sponsored content on Facebook is not positively associated with future purchase frequency.
  4. Of the measured emoticons, only the Love emoticon showed any statistically significant relationship with post-engagement purchase visits. Beyond Love there is likely to be little value in measuring customer engagement via emoticons, except perhaps under special circumstances, such as a humorous Facebook ad campaign.
  5. The study found empirical support for the general hypothesis that engagements reflecting strong DCE are more firmly associated with purchase frequency than those reflecting moderate or weak DCE. 
Featured in this Story
Timothy Halloran
Principal Lecturer

This website uses cookies. For more information review our Cookie Policy