Right, do me a favour. Go on – it’ll only take a second. Look at your hand. Either one, doesn’t matter which. Now, I want you to think about what is good about that hand, and also what’s bad about it. Look at it in real detail – look at the nails, the cuticles, the knuckles. Look at your palm and all the little wrinkles and bumps and freckles. And think about which of those things are good, and which of them are bad.
If you haven’t done it, I really urge you to. It’ll help me make a point in a minute and I know you want to help me do that. If you have – thanks! I hope you didn’t look too weird on the train. I’ll play along and do it too:
Looking at my hand, I can see that there’s a scar on the back of it – that’s bad. My nails are looking pretty good at the moment though, so we’ll chalk that up as a plus, but I’ve been gnawing at the cuticles so they’re a bit red – bad. They’re also not quite the picture of youthfulness they used to be so I’m going to stop looking at them now, I’m getting a bit depressed.
So, here’s my point. 99% of the time, I don’t notice any of those things. That’s because the things I need my hands for the most – holding onto things, typing, throwing gang signs (ok, maybe not) and so on – none of these things are impacted in any way by the condition of my nails and cuticles, nor are they affected by a scar I got ten years ago in a dodgy bar in Staines. However, at some of the places I’ve worked, these would be major issues. The team would insist that, because the user pointed out these flaws, we must fix them before continuing.
I particularly faced this issue once when working on the design of a public service. I was the Lead User Researcher, planning and conducting research while also upskilling other members of the team who were interested in the role. What was great about that team was that everyone made a real effort to attend the user testing, which was conducted in a very nice research lab with a live feed to an observation room. The problem was, they found it difficult to choose their battles.
One of the problems with conducting any user research that the user’s actually aware of is that they feel they must give you their money’s worth. They will pick apart a design, and look much more closely at your website, app or service than they would when using it in real life. This is why, in user research, you’ll so often hear things like,
“It’s alright for me but I think other people wouldn’t like X”
“Someone like my mum would probably struggle with that”
“I think people might have a problem with Y”
Your users are saying these things because they have no other input of their own to add – they are truly grasping at straws so that they don’t seem like a waste of a test. That’s why I always follow up on these remarks by asking if that same problem affects the user you’re speaking to. If it doesn’t, it’s probably not worth noting.
Where it gets more difficult is when the user attending your research session points out something as their own feedback, but you know from experience that they’re only saying this to give you something to write down. This is really only something you notice after conducting or attending a lot of these sessions, and having a good understanding of how users think and behave; that’s why so many UXers come from a background in psychology. And because of that, I can understand when a team without a UX background immediately wants to jump on everything that users have said and start prioritising changes.
With that in mind, here’s some things to consider before you act upon user feedback:
So what can we conclude from all this, now I’ve told you my million dollar secrets – that are not secrets, nor are they worth a million dollars? The conclusion is that while anyone can run user research, and anyone can interpret the results, if you have brought a User Experience Designer, User Researcher, or something similar on board, please listen to their advice. They’ve seen a whole lot of these sessions, and they know what they’re looking for.
A user pointing out something that’s wrong with your service doesn’t necessarily mean you need to fix it.
Major thanks for the post. Cool.