Menu

Research

Why asking users what they want rarely works

Users are great at reacting to what's in front of them. They're much less reliable when asked to design the solution.

Why direct user suggestions often solve the wrong problem, and why the real insight usually sits underneath what people ask for.

09 March 20254 min read

In short

Why direct user suggestions often solve the wrong problem, and why the real insight usually sits underneath what people ask for.

Why direct answers can be misleading

So we did exactly that. We asked the user what they'd expect to happen next, what they'd change, what would make the journey better. They gave solid, clear, confident answers that, read back in isolation, would sound like useful direction. The problem was, when you stepped back and looked at the journey, none of those answers really solved what was going on.

Users are incredibly good at reacting to what's in front of them. They can tell you when something feels off, when something is frustrating, when something doesn't quite make sense. But when you ask them to design the solution, even indirectly, they're working from a completely different vantage point. They're solving the moment they're in, not the as a whole.

None of this makes the answers worthless. What somebody asks for is reliable evidence that something is wrong at that point in the journey. It is unreliable as a description of what would fix it, and the two get conflated because the arrives sounding like a specification.

Users are solving the moment they're in, not the system as a whole.

When user requests point to the wrong fix

On a travel , this became obvious quickly. Users were going through a booking journey where the early stages felt smooth and exploratory. But as they got closer to committing, hesitation crept in, not always verbally, but in how they behaved. They slowed down. They re-read things. They started second-guessing decisions they'd already made.

When we asked them what they wanted at that point, the answer was almost always the same: more information, more , more visibility of pricing and options earlier in the journey. It sounds completely logical. But when we actually explored that direction, the effect was the opposite of what they expected. Bringing more information forward didn't increase . It introduced more doubt. Users started comparing options they didn't need to compare yet, questioning decisions before they had enough , and ultimately dropping out earlier. What they had asked for made the experience heavier, not better.

What they were really telling us wasn't we need more information. It was we don't feel comfortable committing yet. Those are two very different problems.

The general shape is worth recognising because it recurs constantly. A for more of something is usually a request for reassurance, and reassurance is rarely delivered by volume. It comes from knowing what happens next, what it will cost, and whether the decision can be undone.

Key takeaway

What users ask for is often an expression of discomfort, not the actual solution to the problem.

Why this gets more obvious in complex systems

Across NHS work, users would often say they wanted everything in one place: all the information, all the options, all the answers visible without having to navigate around. Again, it sounds right. But when you start designing for that, it very quickly becomes overwhelming. What users actually needed wasn't more access. It was better guidance. They needed to feel like they were being led through something, not dropped into it. That distinction only becomes clear when you stop focusing on what users say and start paying closer attention to what they do.

Everything in one place is probably the most common of it. Taken literally it produces a page nobody can scan. Taken as a signal, it usually means people have been made to hunt across several places and do not that they have found everything.

The distinction between access and guidance is the useful one to hold on to. Users asking for more access are frequently describing a structure that gave them no sense of where they were, and the fix is sequence and rather than volume.

What good research does instead

The job isn't to take what users say and translate it directly into . It's to understand what sits underneath it. When someone says they want more control, it might mean they don't the . When they ask for more options, it might mean they're unsure they're making the right choice. When they say something is confusing, it might not be the interface at all. It might be how the journey is structured.

Those are very different problems, and they require a different level of thinking to solve. So I don't ignore what users say. But I don't treat it as the answer either. Because the goal isn't to what users ask for in the moment. It's to build something that works for them when they're actually using it. And those two things, more often than not, are not the same.

The habit that makes this work is asking what happened just before the . Somebody wanting more control has usually just experienced something they could not predict or undo, and that event is far more actionable than the they proposed in to it.

Written by Andy Scott

Strategic design, UX and digital transformation thinking from real projects.

LET'S WORK TOGETHER

Ready to improve your product?

UX, research and product leadership for teams tackling complex digital services.

Previous feedback

I had a fantastic experience working with Andy. One of his most impressive achievements during our time at NHS HEE was masterminding a deeply complex information architecture for a new platform that brought together a large number of legacy websites.

Will Parkhouse

Senior Content Designer