USER EXPERIENCE DESIGN
Using evidence and instinct in design
You don’t need to prove the benefit of every design decision you take. But you’re not a psychic design super-hero, so use evidence where it exists.
I was recently reminded of one of the many arguments which seem to polarise design discussion online. The reminder came in the form of a Medium post titled Data-driven design is killing our instincts.
In the post, the author argues that an over-reliance on data in favour of design instinct is harming users. I understand where he is coming from on the data-driven point, but don’t accept that we should defer to instinct where evidence exists. To some extent it seems he is looking for a free reign to take decisions without question.
The need to prove every little decision before taking it, is not an effective approach. But neither is betting everything on the instincts of a designer.
Data is just evidence
When people use the term data, they tend to mean numbers, be it analytical data, the results of A/B tests or whatever.
The numerical data you have about your users is evidence of their behaviour and this can and should be used to inform decisions. However it’s just one source of evidence and you can’t rely on it alone. When you do, you tend to layer badly formed assumptions on top of facts and then treat the assumptions as though they too are facts.
A growing example of this comes from some teams who use session capturing software such as HotJar or Full Story. I have worked with teams who initially believed this to be ‘more reliable’ than carrying out qualitative research such as usability tests. But when they did both, they quickly learned things they wouldn’t have otherwise understood. They also notice that some of their ‘facts’ were just guesses.
I recommend you (ethically) grab as much evidence as you can about your users, from as wide a range of sources as possible. This will build a detailed base of evidence which can be referred to when making a wide range of decisions.
Relying on a single source is unwise. As is confusing evidence with proof, so let’s talk about that next.
Confidence over validation
Validation happens only in the live environment, when your software has been released into the wild.
The evidence you use to inform your design will never be proof of the decisions you should take. It’s not even helpful to think about it that way. Evidence will only ever guide you and build confidence in your decisions. That evidence is generally a better guide than anyone’s intuition.
I frequently challenge the teams I work with, to refer back to the existing evidence we have when making decisions. The evidence is sometimes pointing one way when everyone’s intuition is pointing the other.
I’d love to say that this advice is always listened to, but in many cases a few failed attempts are necessary before the penny drops.
The place for instinct
Designers take thousands of decisions with every design they create. They can’t inform every single one of those decisions with evidence and this is where intuition or instinct comes in. Your instincts are just your best guess based on past experience.
You use current evidence to inform your design decisions and where it can’t help, you make assumptions. We then test those assumptions, with users, not for validation but as an opportunity to spot the most important ones we got wrong.
Feeding instinct
I’ll say again that your instincts are just your best guesses based on past experience. That experience can’t come from the designs you have created in the past, nor the design crits you’ve had with colleagues. Instead it needs to come from watching target users trying to use your designs.
Designers who seek to go unquestioned, tend to never see and understand the flaws in their instincts. As a result those instincts don’t really improve over time. In my experience, such designers tend to have instincts which are no better than their non-designer colleagues. They don’t see people struggling with their designs and aren’t refining their instinct for what does and doesn’t work. Their instinct is based on trends and best-practice they’ve read about online (which gave us such successes as the hamburger menu and the homepage carousel).
In contrast, those designers who seek to pick holes in their work and understand the weaknesses in their decisions, tend to develop the best instinct for making good decisions. This is because they receive the kind of feedback which helps them to improve it.
Inform risky assumptions
In conclusion, my advice is to inform your decisions by building up as much evidence as you can about your users and their context. But understand that you will rarely have proof of anything. Instinct has a place, but its place is where there is no reliable evidence to call upon.
Find the gaps in your evidence which are causing you to make risky assumptions and plug them up with evidence.
Test your designs to find the important errors in your instincts and watch those tests so your instincts can improve.
Read some other stuff I wrote
If you liked this post then read some more…
Design
- The UX of stamping loyalty cards
- The power of the big green tick
- Design guidelines for mobile date-pickers
- Getting real about delightful design
User research
- Cognitive interviews for user research
- Moderating user research with Zoom
- Communicating UX issues with comics
- Understanding sampling bias
- Usability test tasks to avoid
Product management
