Behavioural science in the design process

Published by Nick Murphy on

Why you should be using behavioural science in your design process

Table of contents

People click your links. People choose your offers. People are attracted to your services. People decide whether what you offer helps them to solve their problems. At the heart of these statements is people, the way people think, the way they behave and the way they make decisions. That’s why you need to consider using behavioural science in your design process. 

It’s true that you might know a lot about your users, audience or buyers —this knowledge has likely been gained through a mix of trial and error, whether on a large or smaller scale, and other potentially more systematic approaches, some A/B testing or something similar. 

You can go a long way with these approaches. The risk with these is that they keep you tightly bound to your current customers, and their current state of mind — as they change you respond to that change with reactions of your own.

What if…Questions for your design process

What if you could understand more of the whole person (or people) with whom you engage  how would that help you design differently? What if you could develop a clearer idea of their future needs? What if you could build the product or service or features they need tomorrow, not the one they bought yesterday? How might you attract new users who are different to your current users?

close up of design team at table designing digital interactions, with abstract digital screens above

When you ask questions like these and you don’t quite know how to answer them, or you find them a little intimidating, that’s when you should consider using behavioural science in your design process. Using behavioural science in your design process is about really understanding your buyer, audience or users’ world and the problems they face. You can then systematically evaluate how their world-construct is informed to help them make decisions in their interest.

The value of an ethical model of design

A little side note, it’s important to come at it from this angle because good behavioural design is about helping your users do things that improve their lives, it’s not about manipulating people to make choices in your interest. It’s important to re-site yourself in the service of your users, their world and their needs. After all, you’re not designing for yourself, you’re designing for other people.

The models of design

When you have questions like these and you want to find new ways to engage your users, where do you start with a behavioural scientific approach? There are several models you can use, some deeper and more systematic than others. 

When we are consulting with our customers about this, we think that deeper is better but there are plenty of reasons why you might not use a super deep model: time considerations, effort considerations, access considerations, for example. So, starting with the deepest, here are some of the models you could be using:

Diagnostic models of behaviour change:

  1. COM-B
  2. The behaviour change wheel
  3. The theoretical domains framework (TDF)

We categorise the behaviour change wheel COM-B and TDF as diagnostic models of behaviour change. These are systems advanced significantly by UCL’s Centre for Behaviour Change as well as other academic bodies. They are the deepest models of behaviour, what we’d call diagnostic models of behaviour, because they provide a systematic approach for understanding what a behaviour is, how it is influenced by the community a person lives in, their environment and their cognitive decision-making processes. This is the type of approach to take when you don’t know why someone is doing something they “shouldn’t” or you’d like to understand how a person thinks about a behaviour.

Design models of behaviour change:

  1. B=MAP/B=MAT
  2. EAST

These design models of behaviour change are categorised differently because they tend to be better at helping when you know what behaviour change you want to make but need some support in designing an appropriate intervention. They tend to involve a rule of thumb that can help you to understand whether the design you have concepted has a higher chance of succeeding or not.

What else do you need for a more behavioural scientific approach in your design process?

Outside of these models of behaviour change, there are a number of critical things you need to be doing to ensure you get the best from behavioural science in your design process.

  1. Research using tools like surveys, interviews and focus groups – the more these resemble a behavioural diagnostic question set, the better
  2. Behavioural research – how have other people understood and documented similar behavioural challenges?
  3. Behavioural design research – what interventions have other people used to solve similar problems before?

By doing these things you don’t guarantee a successful design, but you do have a much better chance of designing in a way that ensures you can get insight if it doesn’t work. Being systematic means you can identify errors in your own judgment, challenge your own thinking, and articulate why something you’ve designed has worked. Developing a holistic approach has huge potential for your design processes and could be the approach you’ve been missing to get you out of a design rut.

So, although there may not be any guarantees from using a behavioural scientific approach to your design process, it’s  one (giant) step closer  to a better-informed design process.

abstract digital design connections

Related stories

digital user experience conceptual image
Best Practice

The inaccessibility of dark patterns

Even if you haven’t heard the phrase Dark Patterns, the chances are you’ve almost certainly come across them — both online and in the physical world.

conceptual image of AI
Best Practice

Do androids dream?

Love it or loathe it, AI technologies already interact with many aspects of our lives. So what have we learned about attitudes towards AI?