How Brand Social Posts Move Sales

ContentLiadeli, G.; Sotgiu, F.; Verlegh, P.W.J. · 2023Journal of Marketing
Topicsowned social media·social media engagement·sales elasticity·meta-analysis·content strategy·functional vs hedonic content

Imagine you are approving next year's brand social plan. It tracks followers and engagement, meaning reactions such as likes and comments and sometimes their positivity, but not sales. Do you judge brand social posts on engagement alone, or on sales too?

On average, brand social posts are associated with a stronger response in sales than in engagement.

Data chart

Sales response is stronger than engagement response on average

Average percent change in sales and in social media engagement per 1% increase in owned social media, measured as post volume, positivity or presence, across 86 studies. Engagement means reaction counts and, in some studies, their positivity.

+0.35%Sales response+0.14%Engagement response
  • Measure sales alongside likes, comments and shares.

Judge brand posts on sales as well as engagement.

Table 2 (p. 417): exact overall averages .353 sales and .137 engagement, pooling post volume, positivity and presence. These are not posting-frequency estimates; volume-only means are .345 sales and .198 engagement. Outcomes are measured at different levels and are not directly comparable (p. 415). Predicted sales response falls from .616 to .046 at the 25th and 75th percentiles of data-collection year, with other study features at their means (Table 3, pp. 418 to 419; data 2007 to 2019).

Data chart

Informational content shows a larger sales response than emotional content

Predicted percent change in sales per 1% increase in owned social media, by content type, with other study features held at their averages. Owned social media is measured as post volume, positivity or presence.

+0.58%Informational content-0.07%Emotional content, not significant
  • Informational content shows the larger sales response.

Lead with informational content when your goal is sales.

Table 3 (p. 418), as sketched in Table 5 (p. 422). Exact predicted sales responses: informational .580 (p < .01), emotional -.073 (not significant). The content-type difference is significant at p < .05 (p. 416). Other study features, including measurement indicators, are held at their sample means. Informational and social sales responses are not reliably different; this comparison does not establish informational content as best overall.

Action guide

  1. Judge brand social posts on sales, not just easily counted reactions.Sales show the larger average response.
  2. Match your content to your sales or engagement goal.Lead with emotional content for engagement and informational content for sales, and expect little from deals.
  3. Prioritize community quality over follower growth.Smaller communities show the larger relative sales response.
  4. Emphasize new products in your brand social content.Their sales response is roughly six times that of mature products.
  5. Weight your platform mix toward social networks for engagement.Engagement responses are weaker on microblogs.

Evidence

  • Across pooled studies, a 1% increase in owned social media, measured as post volume, positivity or presence, is associated with about 0.35% more sales and 0.14% more engagement.
  • Both averages are positive, yet about 37% of engagement estimates and 16% of sales estimates are negative.
  • Emotional content is strongest for engagement; social content such as questions or contests is weaker. Informational content has the highest sales estimate, not reliably above social content. Deals have the lowest engagement estimate and no reliable sales effect.
  • Smaller communities show the larger relative sales response: with other study features at their averages, a 1% increase in owned social media predicts about 0.41% more sales, against 0.27% for larger communities.
  • New products show roughly six times the sales response of mature products.
  • Social networks show stronger engagement responses than microblogs.

Key takeaway

Brand social posts are associated with a stronger average response in sales than in engagement.

Limitations

The studies cover 2007 to 2019 and measure sales and engagement, not profit, at different levels, which limits direct comparison. Across study years, the predicted sales response per 1% increase in owned social media falls from about 0.62% to 0.05%, no longer a reliable effect.

Source

Liadeli, G., Sotgiu, F., & Verlegh, P. W. J. (2023). A meta-analysis of the effects of brands' owned social media on social media engagement and sales. Journal of Marketing, 87(3), 406–427. https://doi.org/10.1177/00222429221123250

https://doi.org/10.1177/00222429221123250

Exact values behind the rounded figures: .353% sales and .137% engagement per 1% increase in owned social media measured as post volume, positivity or presence; predicted sales responses per 1% increase in that measure, with other study features at their means, of .411% versus .268% for smaller versus larger communities, .650% versus .104% for new versus mature products, and .616% versus .046% at the 25th versus 75th percentiles of data-collection year; informational .580% and emotional -.073% predicted sales responses on the same basis.

Read the paper ↗