Designing for behaviour change is a method for using cognitive biases and motivations of users to design better and more efficient products.
Let’s get theory out of the way
The Dual Process theory
The Dual Process theory describes how thoughts come into being in two very different ways.
System 1 – Implicit thinking
This is our unconscious reasoning which performs intuitive, low effort thinking. It uses heuristics (shortcuts based on prior experience) to do things.
System 2 – Explicit thinking
This is our conscious rational reasoning that performs slower and logical thinking. It needs more focused attention to get things done.
System 1 versus system 2
During our daily routines, we are not consciously thinking of what to do next. We often act on habits and we base these habits on prior experience. Using system two to get things done is hard work as it takes attention and cognitive power to use.
When we are designing for the user, we want to target system 1, which is the automatic part. This puts fewer obstacles in the user’s way. Alright, we are done here.
Step one: understand your users
Get to know user’s needs
We start of this process by gaining an understanding of how our users decide.
Questions to ask about your users
- Why aren’t users already doing something in the product? What is hindering them?
- Where is there an opportunity to change in people’s behaviours?
Build on these questions
- Interview users about their work life. Find out what makes them tick in relation to the jobs they want to get done.
- Build job stories from what you learn e.g. “Tom is a busy manager, and he wants to report on company sales, so he can present to the board.”
Step two: model behaviour
BJ Fogg’s Behavior Model
The Behavior Scientist, BJ Fogg shows that there must be three elements for behaviour to happen:
- Motivation: When there is an amplification in the intrinsic and extrinsic motivators
- Ability: When the behaviour is easy to perform.
- Trigger: When there is an increase in the number of triggers
Now let’s extend and refine Fogg’s model to be more useful, basing it on a model proposed by Stephen Wendel.
If the person passes through these five stages, then there is a high probability that they will execute the action. If the action is a habit, then evaluation and timing are redundant.
The trigger can be internal or external. External cues happen when there is something in our environment triggering us to think about a certain action. Internal cues result from our minds thinking about the action on its own.
Once the user’s mind has been triggered to think about a potential action, there is an automatic reaction in response. This reaction can be intuitive and automatic.
After an initial intuitive response, there might be room for a more conscious evaluation of the action and of potential alternatives. This happens especially when we are facing novel situations. If habits have formed, they might not need to evaluate.
Assuming they have made a choice to act, the question arises whether it is feasible to undertake the action. The individual must be able to act at once and without obstacles.
When should you take the action? We can take action based on a sense of urgency, and by other, less forceful factors.
Step three: strategies to use
1. Automate things
The easiest and best strategy is to eliminate the work required of the user beyond giving consent, then do it. We can use smart defaults, machine learning and tailoring.
Three technical solutions we can use:
- Automating the action behind the scenes
- Using intelligent defaults
- Making the behaviour a side effect of something else the user is doing
2. Make or change user’s habits
Avoid conscious effort on the part of the user. If the user needs to take an action multiple times, and you can identify a clear cue, routine, and reward, then use the “habits” strategy. These need an immediate reward. Habits are immensely powerful ways to lock in repeated behaviours.
Support user’s conscious action
The last and least is using conscious action. If either of the first two strategies are not available, then you must help the user consciously undertake the target action. This is the hardest.
Step four: reward users
Jeffrey Nevid says, ‘motives are the ‘whys’ of behaviour — the needs or wants that drive behavior and explain what we do. We don’t observe a motive; rather, we infer that one exists based on the behaviour we observe.’
Let’s look at two types of motivation.
“If I sign up to this product, I will receive some rewards in their loyalty program and get something free next time.”
These motives can come from family, professional environment, competitions, or contests. The motives of this sort encourage the person to deal with the outer world of other people as there is no other way to achieve the particular goals.
“If I buy this product, I can help disadvantaged people as some of my money goes to charity.”
Intrinsic motivation creates its own rewards, and they are formed through the wishes and needs of the person. It has meaning. It is not a magic bullet, but intrinsic is the most powerful motivation to use.
Helping people focus on the meaning and impact of their work, rather than on, say, the financial returns it will bring, may be the best way to improve not only the quality of their work but also–counterintuitive though it may seem–their financial success. ~ Amy Wrzesniewski
Step five: choosing the actions
Develop job/user stories
Write out what goals the user wants to achieve.
Generate a list of possible actions
List the specific behaviours we want the user to do. Who is doing it? What is the person trying to achieve?
Choose the best action
After a group discussion, select the best action based on job/user stories.
Structure the action
This can be a simple flowchart or a user journey. Think about the sequence of real world steps a user needs to take to complete an action.
- Are these steps easy and intuitive to perform?
- Where might the pain points in each step?
- Are they meaningful enough to reward after completion?
- How does the user finish each action and how do they know? Is there any feedback do they get?
Design the environment
To support the action, an appropriate environment needs to be created.
- Ensure your users are motivated to act, and that motivation is at the front of their minds. Either accent their own existing motivation (best), or motivate them with money, approval, social status, or other goods
- Ensure they are cued to act now. Easy version: ask them
- Ensure they know they are succeeding or failing—give them actionable feedback
- Avoid or co-opt other behaviours that are competing for their attention
- Ideally, piggyback onto something they are already doing
Step six: putting what you know in to practice
Job stories can come straight from the behavioural plan. Each step in the sequence of actions the user takes building up to the target action gets a job story.
“As a manager, I want to be reminded to prepare my company’s sales report, so that I am ready for the board presentation.”
In this example, we know that the manager is under pressure and busy. He wants to be prepared for those big board meetings. What we can do is automate the process of gathering reports. We can remind him, generate those reports automatically, and send them in a suitable format that he can use straight away.
Psychological strategies to use for each stage
- Tell the user what the action is
- Make it clear
- Remove distractions
- Engender trust by making your design aesthetically pleasing
- Engender trust by using social proof
- Engender trust by using a strong authority on subject
- Be personal
- Prime user relevant associations
- Use peer comparisons
- Use competition
- Avoid cognitive overhead
- Avoid choice overload
- Give timely feedback
- Smart default as much as possible
- Use AI
- Use positive peer comparison for easy action
- Use progressive disclosure
- Frame text to be relevant in that moment
Step seven: testing and evaluation
After putting in all this hard work, now is time to test that it is working.
Measure the impact of the product, based on clear target outcomes and well-defined metrics for each outcome. Here it is important to set clear thresholds for success and failure.
Identifying obstacles to behaviour change
Discover problems, develop potential solutions and generate additional ideas for how to make the product better. One can start this process by watching real people using the product (direct observation) and by collecting usage data. We can thus start getting a better insight into how people use the product, what the bottlenecks are, where the product is having the most impact on people, etc.
Learning and refining the product
Determine what changes to implement through:
- Gathering lessons learned and potential product improvements
- Prioritising potential improvements based on business considerations and behavioural impact
- Integrating potential improvements into the part of the product development process