The research, carried out with "emotional intelligence" software company Realeyes, involved 149 ads across 35 brands, with 22,334 people in six countries.
Using the webcams of participants, Realeyes measured their facial expressions as they watched the ads.
Realeyes’ technology measures the micro-measurements of the face and uses computer vision and machine learning to analyse them, focusing on expressions of happiness, surprise, confusion, disgust and engagement.
The data from this testing was then analysed against Mars’ existing sales lift data for each ad – creating what the partners claim is the largest emotional dataset linked to real business outcomes currently in existence.
The analysis found that the emotions data could be used to correctly identify whether the ads tested had a "no to low" or "high" impact on sales 75% of the time.
Mihkel Jäätma, chief executive of Realeyes, said: "We wanted to see if our emotion measurement technology could distinguish between high and low performing ads.
"Being able to identify strong creative with high sales impact enables advertisers to push these ads, and avoid putting media spend behind those with low, or worse – no sales impact. It’s about spending campaign budgets more effectively, optimising ad creation and media buying at no additional cost."