How do consumers respond to television ads that they have already seen, either on Facebook or on TV?
Neuro-Insight divided a group of 100 U.S. Facebook users between the ages of 21 and 54 into two groups. One group watched a TV show with ads on the first day, while the other browsed their Facebook News Feeds. On the second day, both groups watched the same ads during a TV show.
Facebook IQ offered more details on the study:
During the test, participants in each group wore EEG caps that measured responses correlated with real behavior from different parts of their brains. Neuro-Insight was able to measure their responses against neural metrics approach/withdrawal—the brain’s response to positive and negative emotion—and three key neural metrics that were the focus of our study:
- Engagement, an indicator of how involved people are.
- Emotional intensity, a measure of the strength of emotion being experienced.
- Memory encoding, the rate at which the brain is storing the current experience about the brand into long-term memory, where it can be recalled more than a few minutes later. This metric is recorded for the left brain as memory encoding detail and for the right brain as memory encoding global.
Neuro-Insight found that participants who were primed with the TV ad performed below the 50th percentile for memory encoding, and were more likely to make purchase decisions after viewing the ad on day two.
Meanwhile, participants who were primed with the ad on Facebook scored above average for memory encoding, with the Facebook ad acting as a teaser, similar to a movie trailer.
Neuro-Insight also used two different types of ads during testing: video ads originally produced for TV, and video ads that were optimized for Facebook, with the latter shorter in length and including branding early on. Facebook IQ described its findings:
We analyzed the peak branding moments for each video ad when viewed on TV on day two. When participants were primed with the optimized videos on Facebook, Neuro-Insight found that these optimized ads were more strongly associated with the brand than the repurposed TV ads, producing the greatest change in the memory encoding metric. The strong branding in the optimized video ads was able to increase the brand impact of the later ad exposure when seen on another screen. Participants were more likely to see those branded moments as they browsed through their Facebook News Feed when they were presented early in the creative. Branded moments triggered a stronger reaction when the participants saw the full ads on TV on day two.
Finally, Facebook IQ provided the following bullet points for marketers:
- Reinforce impact across screen: Multiscreening has made it important for marketers to understand how best to coordinate their messaging across channels. We saw within our research that cross-channel advertising was a highly effective strategy for priming. By running a Facebook ad then a TV ad, marketers can help increase the impact of their campaign across screens.
- Maximize mobile’s appeal: Mobile is a compelling advertising platform for marketers. In previously published neuromarketing research commissioned by Facebook, we found that, overall, people were more attentive and tended to feel more positively toward information presented on a smartphone than on a TV. Within this study, cross-channel campaigns first primed with Facebook ads on mobile outperformed TV-only campaigns, showcasing the benefit of including mobile in a campaign plan.
- Design with the screen in mind: When designing video creative, marketers should be mindful of where that creative will be viewed. Develop creative for mobile that maintains the essence of an original TV ad by using short clips or imagery from the key brand moments of a TV ad—this could also help prime people for the full TV ad. If the TV ad features related vignettes, marketers could use each vignette as a separate digital asset, creating mobile teasers that link to the overall campaign and ultimately tell a cohesive story across screens.
Readers: What did you think of the results of this study?