Back to home | Ir a home

Segmentation by psychology: Using Twitter to unearth audience’s personality insights


In the future, every decision that mankind makes is going to be informed by a cognitive system like Watson

Ginni Rometty, Chairman & CEO, IBM

An increasing ability to predict human behaviour will be a growing expectation of digital marketers in the near future, as the possibilities for bridging the gap between digital touch points and the store become more commonplace. With an open network, Twitter is a perfect platform to utilise if you want to research and target your audience on a psychological level. To help digital professionals get a deeper understanding of how to implement the necessary cognitive computing technology into their strategies, Audiense were invited to talk at a Twitter event last week that looked at what Twitter data is telling marketers this holiday season. If you weren’t at the talk, or you want to refresh your memory, here are some of the key insights on how you can use Twitter for discovering personality insights.

How can personality insights help?


Know your audience

Marketers who have a clear image of their audience will be in the best position to produce campaigns that connect with potential customers, and bring them through the door. In the past, the top digital marketing teams have been gathering these insights for audience research, list building, and actionable strategies:

  • Who are your current audience, your competitor’s audience, and your customers
  • What particular groups of Twitter users have engaged with, what they’re interested in, and what they’re actually Tweeting about
  • How certain audiences have responded, the sentiment of those responses, and when your audience are online

These will still be vital going forward, but there is now another layer that can be added. Personality Insights by IBM Watson can add a crucial dimension to these insights. It uses the Big Five personality model, Values, and Needs to identify a large range of personality traits which can provide marketers with nuanced data on any segment of Twitter users they want to target or analyse. It can add to the above and tell you the following:
  • Why will they respond?
  • What will their reaction be?
  • What personality traits are predominant among a selected group of Twitter users

Optimise & expand campaigns

To ensure your advertising spend is at its most efficient, you can aim your campaign at the Twitter users who are most likely to respond in a manner that aligns with your KPIs. For example, if you want to drive footfall to a store then you can target Twitter users who are more likely to redeem coupons. Perhaps you have a look book that you want to distribute to 10,000 influential profiles who Tweet about fashion, you can identify the ones who are twice as likely to click on it when it’s advertised to them.

The technology can be used to help segment campaigns into specific personas. This will allow you to customise specific Promoted Tweets which align with both the traits of their personality, and the actions they’re more likely to perform.

Assist in defining strategic insight

The data available can give clearer pictures on the audiences of key competitors, specific demographics, and people who interacted with your campaigns. These can be used as data points to assist with your campaign messages, how you structure your Twitter campaigns, and also to identify distinctive features that can assist in how you communicate with target personas.

What are the results?

In a series of tests where IBM Watson analysed Tweets and the actions of their users, several key observations about the activities of certain personality types were noted. Identifying these traits, or targeting by them, will give you guidance on how to extract the optimum results from your campaigns. Some of the results were:

  • Twitter users who are high on orderliness, self-discipline and cautiousness and low on immoderation are 40% more likely to redeem coupons.
  • Individuals with high levels of openness and low emotional range doubled the clicks on targeted advertising.
  • Facets such as excitement-seeking, friendliness, activity level, agreeableness identified the people who were most likely to respond to Tweets.
  • If you are targeting professionals, a higher self-transcendence value will be more likely to read articles related to work.

What to do next

The understanding and implementation of these techniques can be greatly improved by developing an understanding of The Big Five Personality Model, a psychological profiling system that IBM Watson uses. To learn why this technology is vital to marketers, check out Audiense’s webinar designed specifically for advertisers and learn key insights such as:

  • How does the model predict human behaviour and what does it mean to advertisers?
  • What is cognitive computing?
  • How psychology driven marketing can predict an audience's brand preference, likeliness to respond to ads and redeem coupons.
  • How you can get started with psychology-driven marketing on Twitter

Want to implement these strategies into your campaign? Learn all of this and more in our exclusive free webinar with IBM, or contact us to arrange a demo.

Twitter Marketing - SocialBro