MENU

Advertisers: How Cognitive Computing Will Change Everything

IBM Watson Advertising on Twitter - SocialBro Audiense

What personality traits do your most profitable customers have? How do they differ from your competitors’? If only you knew, you could dramatically optimise the way you communicate and win more hearts and minds, right?

We are entering a new era of computing and it’s giving us answers to questions like this. Cognitive computing systems learn and interact naturally with people. This is transforming the way we understand human behaviour. Today, Audiense announces its cognitive computing offering for marketers.

We’re incredibly excited to tell you that we have integrated IBM Watson’s Personality Insights API into our platform. This means that you can now use Audiense to get a detailed personality profile of Twitter users. You can analyse personality traits, needs and values, all based on the language they’ve used on Twitter.

Once you know (or have hypothesised) the personality of a group you want to target, you can then use Audiense to find that audience on Twitter based on any of those traits. This can also be combined with Audiense’s social graph, content, and demographic search. The traits include ‘more likely to propagate information’, ‘more likely to redeem a coupon’ and ‘more likely to respond to unsolicited advertising’. The criteria are all based on studies done by IBM Watson and Audiense that show personality can predict consumer behaviour with significant accuracy.

IBM Watson SocialBro Audiense Big Five Conscientiousness

We’ve been beta testing this with a few brands and the results are very promising. In one example, we took all the profiles who were unlikely to click, then segmented them out of a Twitter Ads campaign. Allowing us to focus on targeting those who were 40% more likely to redeem a coupon. This we know because they scored high on the ‘Big Five’ facets orderliness, self-discipline, and cautiousness and low on the facet immoderation.

With another client, we looked at all their most loyal customers who had recently defected to a competitor (using data shared by the client). This allowed us to help the client discover how the personality traits of those defected customers differed from their broader audience. Following this, we could look at how a competitor’s followers differ from the client’s, or who their most adventurous followers are.

With 47 personality traits, needs and values analysed by Personality Insights, the level of insight that you can glean is impressive – and a LOT of fun to play with. There is a lot of learning (and thus competitive advantage) to be gained here. We learnt recently that people who engaged with a video of a comedian in a promoted Tweet campaign we ran recently, were consistently more likely to be highly neurotic, agreeable and extraverted.

With cognitive computing, we see patterns of behaviour and reveal why people do what they do. Once we understand them, we have a real chance of influence. Because of this, we believe that cognitive computing is the the start of a new era in marketing and advertising too. A time where we leverage cognitive computing and artificial intelligence to build meaningful relationships.

This is the best opportunity marketers have ever had to deliver ultra personalised, deeply relevant messaging, accurate targeting and exceptional response rates. Last year there was a lot of discussions in marketing about sales and marketing alignment. With cognitive computing and personality-driven marketing, we align directly with the customer psyche. This is a great time to be a marketer.

New Call-to-action

  • As an avid Watson fan and a big believer in utilising AI for contextual information and of course a Social Bro user this is awesome news!

    • Luella

      Thanks Nat. We are very excited about this launch too!. I’d love to hear how you’re using Personality Insights and your ideas for the next iteration for the product. Just tweet @1uella. Have a great week.

  • mc900

    WOuldn’t Draper kill for ANY analytics coming from the 60’s and 70’s?