Behavioral Design

What is Behavioral Design?

Share This Blog Post

Companies that understand the three foundational principles of behavioral economics know that their customers are not fully rational. They know:

  • We are influenced by the environment of our decision-making.
  • Giving people information doesn’t change behavior.
  • Our intuitions about what actually changes behavior tend to be incorrect.

Companies who understand this use behavioral design.

Behavioral design uses what we know about the psychology of decision-making — behavioral economics — to purposefully inform design decisions. Experience designers use this systematic approach across industries to successfully design products, revise features, and change customers’ behavior for good. In this article we will teach you how to use the behavioral design process in your own organization.

The behavioral design process has three primary phases:

Do a behavioral diagnosis.

Identify your desired behavior and outline every step your users have to take to achieve it.

Identify psychological biases.

Determine the barriers that get in the way of decisions and the benefits that motivate them.


Choose one barrier to address and design an experiment to test an intervention.

1. Behavioral Diagnosis

As behavioral designers we study what people do and not what they say they do. A Behavioral Diagnosis is the tool to do this. It is a literal map of everything your customers actually do.

Where do you start? First, you have to identify a key behavior – the action that you want your customer to complete. It doesn’t have to be the final step in your customer experience; it could be any action you want them to take. If you don’t know your key behavior yet, that’s fine too! The next step of the process will help you figure out where to focus.

Next, you create a behavioral map by outlining each and every step to get to your key behavior. If you don’t know your key behavior yet, map out as many steps of your customer journey as you can. This should be a detailed list of every action a person must take to get to (and ultimately perform) the selected behavior. Once you have the steps laid out, layer on any data you have about how many people complete each step, conversion metrics, and more. With these steps and data identified, select one of the most problematic steps for your customer – where they’re most likely to drop out – and use that as your key behavior.

Example: Imagine we identified the problem area as early childhood math learning in public education, and the key behavior as having a substitute teacher successfully complete full lesson plans. In the mapping stage we would look at the small details of the environment to identify what was happening. For example, the sub teacher comes in 15 minutes prior to the start of class and looks at his phone for 10 of those minutes. He reviews the 5 sentences of instructions left to him by the teacher prior to class starting. He does not acknowledge the kids as they enter. He takes the first 10 minutes of class to get to know students’ names. The class list has errors. The kids correct two of the names. He spends 5 minutes of the 15-minute lesson on directions to ensure the kids understand what’s going on.

By focusing on what people do instead of what they say, we can generate hypotheses about how to solve a problem based on actual behavior.

We ask ourselves: What is each and every step that a person has to complete to successfully reach their behavior?

2. Identify psychological biases

The second step of behavioral design is to identify and label the psychological biases that a user faces as they move along the behavioral flow. The world of psychological biases gets complex very quickly – Wikipedia lists close to 200! To simplify this world, our team uses a model we call the 3Bs. We’ve used the 3B approach with a range of companies to assess and solve some challenging behavioral problems. By no means is this framework inclusive of all the biases, but it does give us a place to start when analyzing a system. Once you master the main concepts of the 3Bs, you’ll be able to go a level deeper and identify the specific psychologies at play.

3. Experiment!

The final step of Behavioral Design is to design a solution that increases the key behavior you picked. To do this, we pick a barrier we want to remove or a benefit we want to add to the system. We then create a hypothesis and design a controlled experiment to test if we change the likelihood that someone will complete our key behavior. We define what we think will happen as a result of our intervention by using the format “If ____, then ____, because ____.” Our hypothesis should help clarify why we are testing this intervention and what effect we expect to see.

What key behavior is your team designing for?

Based on our experience working with clients, this is the most important (and least appreciated) step in behavioral design. While it may sound obvious, this is actually the step teams trip up on most.
Teams are great at assigning outcomes. For example: this project is intended to drive “active product use” or “customer retention.” Such KPIs (key performance indicators) are important for projects.
But project teams need to take it a step further. To identify your key behavior based on KPIs, or identified outcomes, think through:

  • What behavior must change in order to achieve this KPI?
  • What must your customer do?What action do they need to take?
  • What do we want them to do as a result of this KPI?

We recommend you take the following Key Behavior exercise back to your team. Have everyone do it individually and then compare answers. You may think the key behavior is obvious, but answers within teams are often different.

This is where the fun begins.

  • Get uncomfortably specific.
    The more specific the better! Think through who, what, when, where, and how. If I’m designing a meditation app, I’d say: a user starts a 10-minute meditation every morning in their home before they leave for work.” NOT: “A user starts a meditation.”

Pro tip: we often hear from product and marketing teams that they want their customers to have “high engagement.” But it’s your job to get much more specific. What behavior does a user have to do to be ‘engaged’? What does ‘engagement’ actually look like? Engagement itself is not a behavior.

  • Focus on Actions Not Outcomes.

A behavior is an action that someone takes. What do you want your customer to DO?

Product and marketing teams typically measure active usage or retention, and have user logins as a goal. However, these are outcomes, not behaviors. You should ask: What behavior do you want your customers to do after they log in? Logging in itself is not a good key behavior.

An action: “My user will post an update to their community on our platform.”

Not an action: “My user will think the community on our platform is valuable” or “understand how to post.” Thinking and understanding are not actions.

What is Your Key Behavior?

  • Don’t focus on outcomes.
  • Be specific: Specify when a person should do this behavior.
  • How often should they do this? Just once or every day? For how long?
  • Log in and sign up do not qualify as key behaviors.
  • Focus on behaviors. Don’t use words like “They should believe…”

Contact Avatardesk

Contact Avatardesk today for a behavioral diagnosis. Don’t worry, we all have psychological biases! Let’s experiment together to determine how best to specify desired behaviors of your customers, remove barriers from their user experience and amplify the benefits of them adopting your product or service.

More To Explore