Strong emotions lead to strong actions, making emotional analysis an intriguing capability in digital marketing.
In the run-up to the 2016 Presidential election, like most good data companies, audience.ai wanted to make a prediction on the outcome to demonstrate their capability. And just like other tech companies (and the pollsters as it turned out!), they were getting readings that jumped back and forth by the hour. It was too close to call and the data was very inconsistent.
audience.ai specific capability in this area is conversational data analysis. Simply put, that means they look at what is said online from which they can determine the intent, emotions, and sentiments behind those statements. When you have an election that is particularly useful because what someone writes is akin to an unprompted confession — unlike the pollsters who have to ask the question ‘who are you voting for’, audience.ai learns it without asking, therefore getting a more honest insight.
Still, though, 6 weeks to go, the data would not reveal who the ultimate winner was going to be — until the data science team hit on a fascinating idea — to look at negative emotions, not positive. Whilst the LIKEABILITY and LOVE to both Clinton and Trump were too similar to make accurate predictions from, the strength of DISLIKE and HATE was polarizing, and 6 weeks out from election day, they were later proven to have accurately called 49 out of 50 states.
For marketers, strong emotions as a signal of intent might just be the next big thing, and for us marketers, the good news is that there is a crazy amount of emotion projected online, and therefore scaleable data for most targeting criteria.
It’s surprising to me that in 2018 brands and agencies are still forced to work with targeting data that has no real science behind it — ‘auto intender’ segments are often based on one visit to a car auction site, or ‘home buyers’ from being tagged on a property site, even if they are just looking at the value of their own home.
With emotions, we have a much more immediate message.
“I HATE MY CABLE” signals that competing Cable companies and streaming services should be engaging that individual.
“I AM GOING TO GET A BIGGER PLACE” should get realtors, furnishing stores and mortgage advisors frothing at the mouth.
So far audience.ai has been able to match emotions to a wide range of intents. Once a need is identified the data science team will produce customized segments, and upload them to the DMP or DSP of your choice for easy activation.
Every one of you reading this that runs digital marketing programs has a data need that has gone unmet. Try them. The team is very consultative and available on firstname.lastname@example.org.