Social media has had a profound effect on the food industry. In the competitive food media landscape, “Instagrammable” food has become a marketing trend with millions of visually-appealing food images posted on social media each year. Industry trends and conventional social media wisdom suggest that restaurants promote unique appearing foods to stand out so they garner higher engagement (e.g., likes and comments). Think unicorn lattes and Toronto’s own Poop Café.
Yet, using Google Vision AI on a large dataset of Instagram posts from over 850 top restaurants from around the world, our new research evidences that consumers are actually more likely to engagement with foods that appear more typical and normal.
Therefore, despite food industry and social media trends urging restaurants to alter their menus, prioritizing visually uniqueness with their food over taste, all to appear more to a social media audience, this may not actually have the desired effect of increasing social media engagement.
Experimental evidence suggests that in the busy social media landscape, where the average user is exposed to hundreds of posts a day, seeing a food item that is more easily recognized as being normal and familiar, elevates our positive feelings in comparison to a food item that appears atypical and different. These elevated positive feelings then carry over and increase the likelihood that a social media user will hit the ‘like’ button.
Thus, “Instagrammable” food is actually foods that simply appear normal, familiar, and typical. And it is not the foods that appear unique, different, and obscure.
To learn more, see the full article:
Philp, M., Jacobson, J. & Pancer, E. (2022). Predicting social media engagement with computer vision: An Examination of Food Marketing on Instagram, Journal of Business Research. https://doi.org/10.1016/j.jbusres.2022.05.078 (external link, opens in new window) .