Mars, Inc. wanted to evaluate if their advertising evoked the intended emotion of their consumers and if their emotional response to the ad could predict sales.


Using Affdex for Market Research, Mars and Affectiva developed the first and largest research study tying facial reactions and emotional responses to sales effectiveness. Over 1,500 participants from France, Germany, UK & the US viewed over 200 ads, while their facial reactions were captured via webcam, and emotional responses were analyzed. Using single-source data, the emotional responses were tied to the sales volume for each product line: chocolate, gum, pet care and instant foods.



Affdex facial coding and emotion analytics was able to more accurately predict short-term sales than relying on self-report methods alone. However, combining facial coding with the self-report survey provided the best sales prediction results, with an accuracy of 75%.

With Affdex, Mars also found that across product lines, chocolate ads elicited the highest emotional engagement, while food ads elicited the least; and across markets the US was the most expressive while the UK was the least expressive.