top of page

How Predictable are you?

Mark Zuckerberg knows what you put in the ballot box.

Well, he doesn’t, but he could probably take a pretty good guess. A Cambridge University study undertaken in 2013 unveiled that Facebook could not only predict personality traits, but based on data activity by the user including their preferred profile picture, it could also predict political alliance, sexuality and even those more likely to exhibit signs of depression and drug use.

Many of our own behaviours are driven by ‘latent psychological constructs’ which to me and you, means there are mathematical calculations that can help predict how we’ll act across different contexts, in fairly consistent ways, based on one known behaviour trait.

For example: Jessica votes for a liberal political candidate. It is more than likely Jessica will:

  • Be more open to polyamorous relationships

  • Read philosophy

  • Listen to jazz

‘Why?’ you ask…Well that’s because there’s an underlying trait driving all of those behaviours.

This principle is known as ‘thin slicing’. If you were given an incredibly thin slice of cake, you could probably look at it and accurately predict what the rest of the cake looks like. So given a very ‘thin slice’ of a person, it’s possible to make quite accurate predictions about them based on inferred underlying characteristics. This is why first impressions are so powerful.

So, what’s giving the game away?

In a study where participants were shown photographs of people’s faces for just one-

tenth of a second, results found that initial perceptions of personality were fairly accurate. But it’s not just a person’s face that gives away their character. Accurate personality judgements can be made using cues from a photo, like dress sense, a room environment, items around them, location and what they choose to share on their social media profile.

Spotify data analysis found that personality could be predicted not only by genre preferences but also by derived metrics. For example, the trait of being ‘conscientious’ was related to country, soul and funk, as well as having a premium account and being less likely to skip tracks. In a consumer scenario, those same ‘conscientious’ personality types are also more likely to spend on savings accounts and reading, and less likely to spend on takeaways.

With so much of our lives led through social media, smartphone apps and online activity, we can all be read rather well from our digital footprints. That footprint is obtainable by computers that then use algorithms to predict our personality and turn it into valuable targeting data.

Tell me what I want to hear

Putting our marketing hat on, targeting individuals based on inferred personality traits and behavioural interventions are likely to be more effective and have more persuasive susceptibilities. Put a smiling woman on your digital signage and a discount message and you’ll probably see an increase in sales – from extroverted men.

With technologies like footfall profiling, you can track audience gender, age, and clothing, to provide a ‘thin slice’ from which you can infer personality traits. Dress sense is a strong form of expression and a powerful source of intel when it comes profiling your audience’s personality, character and mood.

Once you have identified a prominent personality type, you can then customise your content and designs according to your audience. Here are some examples:


  • Prefer aesthetics to be loud, bright, and warm

  • Respond better to messaging which is positive and social

  • More likely to be nudged by discounting - having things in the here-and-now, sex and humour.


  • Prefer aesthetics that are representational and simple, like landscapes and portraits

  • Prefer messaging to be factual and educational

  • More likely to be nudged by appeals to conformity and achievement

Open to experience

  • Respond better to surrealist, abstract, complex aesthetics

  • Prefer wording to be more loquacious and unique

  • Prefer messaging which encourages them to reflect and come to their own conclusion and which focuses on benefits


  • Prefer aesthetics that are curved, calm, orange

  • Respond well to messaging which encourages them to take other people’s perspectives

  • More likely to be nudged by sticking to morals and avoiding regret


  • Respond best to aesthetics that are structured and romantic and also like dark colours and sad paintings

  • Respond to wording that emphasises feelings of personal distress

  • Are nudged by attractive messengers

In conclusion

Ultimately, being predictable is unavoidable. While we won’t all fit into the metaphorical ‘box’ with all our arms and legs fully inside, it’s clear from a commercial perspective that nudges with the right consideration to personality types have a higher chance of resonating, converting and when done right, can see impressive results.

Pixel ARE WELL VERSED IN profiling and analytic SERVICES, FIND OUT HOW WE CAN HELP YOU here.


bottom of page