UX Diagnostics

There are almost always 10 different ways to create a certain behaviour. That means everything your users do could be caused by something other than what you intuitively think.


It’s time to play the “what could cause that?” game!

Maybe your users are quitting after the first visit. Maybe they don’t spend very long on your site. Maybe they don’t trust you.

There is a good chance you have a theory about the cause of that problem. 

Maybe there isn’t enough content. Maybe it’s hard to navigate to more than one section. Maybe your design is a little rough.

There is also a good chance you’re wrong.

So… what else could it be?

When you have a theory in mind, one of the hardest things is to assume you’re wrong, or come up with other ideas about what could cause the same problems.


UX can sometimes feel like the perfect opportunity to be “right” and that feeling is often wrong.

We all want to be the person with the answers. But it actually shows more intelligence when you realize that there is no way to know the real cause. There are often several realistic explanations for whatever you’re seeing.

Don’t be afraid to say “I don’t know,” or “it could be several things,” to your boss or your client or yourself!

Maybe your users don’t come back because they didn’t like what you made them, or because they forget about your site, or because they only need your service once a year.

Maybe they don’t spend very long on your site because they don’t believe what you say, or because they can’t find the “register” button, or because their friends aren’t there.

Maybe they don’t trust you because they have never heard of you, or because you don’t sell brands they want, or because you’re adding a bunch of extra costs at the end of the checkout flow.

Once you have a list of plausible hypotheses, it’s time to start eliminating them, with science!


UX requires diagnostic thinking.

Like doctors trying to find out why you’re sick, based on your symptoms. You have a cough, and you’re tired, and your anus hurts? Maybe it’s the flu. Maybe it’s a deflated lung. Maybe you were abducted by aliens.

The only way to know is to do tests or experiments that will give you more information.

The flu will have symptoms or treatments or test results that a deflated lung and alien abductions won’t have.

Do the tests. (A/B tests and user tests)

Look for the other symptoms. (ask more questions, get more data)

Try the treatments. (redesign some stuff, see if it helps)

Whatever happens, you will have more information, and more information in UX means fewer ideas that make sense. i.e. — a better diagnosis.

Sometimes you should do an A/B test specifically to prove that you’re favourite theory is right. Or to rule out a possible cause. Or to show a client that their theory is unlikely to be the real reason.

Or maybe they are right! Shit happens! 

Instead of picking one hypothesis and trying to make everyone believe it, try testing instead. When you prove the right answer, you will build respect among your colleagues, you will learn and improve yourself, and most importantly, you will know that aliens are out there, and they do use anal probes.

You will also understand your users better than you did before. if you’re into that sort of thing.


Do some science. Find the answer. Look smart for real.

You might surprise yourself!