To show how effective your designs are, talk about user behavior change
Design isn’t just about looking nice: it’s about changing user behavior
One of my most visually unappealing case studies is often the one that interests interviewers.
It’s a simple, unassuming form re-design that gets people to stop for one reason: I changed the user behavior of 231,000+ field office workers misfiling a particular claim.
When you start to design at scale, one of the most important to understand is the power of designing for behavior change.
Behavior change isn’t something that seems related to the design process.
However, defining the behavior you want to change can be equally important for one key reason: it’s a way to show the impact of design, even if you don’t have metrics.
To understand why, you must first realize we’re not discussing complex behavior change.
Behavior change can be simple, too.
When people hear the words “Behavior change,” they often imagine fields like behavioral psychology, which are about changing complex and drastic behaviors.
However, behavior change for UX is much simpler and smaller in scale. We’re not addressing deep-seated issues. We’re designing to either change undesirable behaviors or create desired behaviors.
For example, imagine you’re asked to re-design a flawed checkout process. A business pulled together a team, a budget, and a timeline to fix the feature because of undesirable user behavior.
The undesirable user behavior, in this case, might be that 65% of all users are abandoning the checkout process. In that case, UX needs to tackle two things:
Figure out Why users behave this way
Figure out What to design that might change this user behavior
How do we do this? By following a slightly altered design process.
Designing for behavior change, or understanding Why
Let’s say our design project is about changing our undesirable user behavior (users abandon the checkout process) to a more desirable one (users complete the checkout process).
We would start our discovery and early design iteration processes to understand Why users behave this way. One of the major research questions would be:
Why are users abandoning the checkout process?
If that’s the case, we may want to watch out for specific behaviors or listen for root causes during interviews. For example, we might note down:
2/5 Users paused for Google Chrome’s ‘auto-complete’ feature to fill in a credit card before typing it in themselves
Users mentioned being uncertain what the final price would be because it wasn’t clear how much shipping was
We’re asking users to get something they might not have on hand (i.e. ‘Serial Number’ or ‘Date of Last Purchase), which causes them to navigate to another tab
etc.
At the end of this process, you should have some answers on Why users might behave a certain way, which should help inform What to Design to fix this.
Dan Winer, Senior UX Designer and strategist, highlights a five-step process on how we might work to change behavior:
1. Design for an emotional response
2. Expect a behavioral response
3. Track the change in behavior
4. Understand the metric your work impacts
5. Comprehend where it fits into the business context
For example, if we need specific information (i.e. Serial Number) for some reason, but it’s causing users to abandon the process, we might ask something like:
How might we get users to gather specific information before starting checkout?
Figuring out design solutions that change user behavior this way can significantly impact.
Behavior change, not ROI, for UX to show value
The “ROI of UX” is an outdated and meaningless term that’s still used because there is a need to show the value of design in business contexts.
There are a ton of issues with trying to provide statements like “My re-design resulted in $100,000 in revenue”, such as:
What Engineers create from your design may be different
What Marketing produces to complement the design has an impact
How Sales pitch your product may contribute a lot
etc.
This is why defining behavior change can be so important. Once you can define the undesirable user behavior, you can use a framework to spell out how UX, Behavior Change, and Product Outcomes are related.
For example:
Users weren’t able to complete the checkout process (Undesirable behavior), which meant we had a 40% checkout completion rate (Product Outcome)
I conducted user research and found that users behaved this way because of X, Y, and Z reasons (User research)
I re-designed the checkout process to address X, Y, and Z reasons (UX Design)
Because of this, users were more likely to complete the checkout process (Behavior Change)
The result was 20% more user purchases because users were more motivated/had less friction while checking out (Product Outcome)
Ideally, you would have a metric that showed the before and after of your design, like “Checkout Completion going from 40% to 60%.” That’s the easiest business impact to show.
However, you can often show the impact of your design and how you changed behavior through other signals, like showcasing a simplified UI or talking about how behavior changed during user testing.
This allows you to show the impact of what you’ve designed in terms businesses understand.
In an age of fancy visuals, lean into the next step.
Right now, design is moving more towards UI. For those who are visual designers, that seems great.
But if you’re like me, who designs things that aren’t always visually impressive (like forms), it can seem complicated to showcase our value.
After all, do people want anything other than high-quality and beautiful visuals?
The answer is yes: Design shouldn’t just look pretty. Design should also drive user behavior change. It’s great if things look nice, but what businesses often care about is driving some desirable action.
Showing that you can provide that to businesses, even if your visuals aren’t the best, can be the difference between an ineffective portfolio and one that gets you hired.
March and April’s Data Informed Design cohorts are now live! March is specifically created for Design professionals in SEA/EA time zones
Kai Wong is a Senior Product Designer and Data and Design newsletter writer. He teaches a course, Data-Informed Design, on becoming a more effective designer using the power of Data.