Menu

IA

Good IA reduces thinking, bad IA creates it

The best information architecture removes the need to stop and work things out. The worst kind quietly adds effort at every step.

Why information architecture is really about reducing cognitive load, and how poor structure makes users think harder than they should.

22 September 20243 min read

In short

Why information architecture is really about reducing cognitive load, and how poor structure makes users think harder than they should.

What good structure feels like

Then there are other times where it's the opposite. You're not stuck, exactly. Nothing is completely broken. But you find yourself pausing. Reading something twice. Hovering for a second longer than you should. Trying to make sure you've understood what's in front of you before committing to the next step. That's thinking. And most of the time, it's unnecessary.

Good removes that burden almost entirely.

The effort is real even when it is small. Every pause is a moment where the user is doing the 's work for it, holding options in their head, comparing labels, deciding what a category probably contains. None of that is what they came to do.

Good information architecture removes unnecessary thinking before users even realise it was there.

Why predictability matters more than neatness

When a is structured well, users don't feel like they're navigating it. They feel like they're progressing through something that makes sense. Each step follows naturally from the last. The way things are grouped feels predictable. The labels don't need interpreting. You don't have to ask yourself is this the right place, because it's already obvious. You're not making decisions about the system. You're focusing on what you came there to do.

IA isn't just about organising content neatly. It's about reducing the amount of thinking required to use something. And when it's done well, it almost disappears.

Predictable and tidy are not the same thing. A structure can be evenly grouped, perfectly balanced and completely wrong, because it reflects how the organisation is arranged rather than what people are trying to do. Neatness is easy to approve in a workshop. only shows up when someone tries to find something.

Key takeaway

Good IA works because users spend less energy understanding the system and more energy completing their task.

What poor IA forces users to do

Bad does the opposite. It forces users to interpret things that should be obvious. They have to work out where something might live. They have to translate internal language into something that makes sense to them. They have to decide between options that aren't clearly distinct. None of that is part of their goal. It's effort created by the . And it adds up quickly.

A single moment of hesitation doesn't seem like much. But when it happens repeatedly across a journey, it starts to change how the whole experience feels. What should be simple becomes tiring. What should be obvious becomes something you have to double-check. drops. And when confidence drops, changes: users slow down, backtrack, abandon things they would have completed if it had felt easier. Not because they can't do it. Because it's asking too much of them.

How you know it is working

The signal that you've got it right is not when users say it's clever, or when say it looks neat. It's when no one has to stop and think about how it works. Because good doesn't demand attention. It quietly removes the need for it.

It is measurable if you want it to be. Give people a task and watch where they click first. First-click accuracy predicts task success better than almost anything else, and it isolates the structure from everything layered on top of it.

The harder signal is what stops happening. Fewer support contacts asking where something lives. Less internal for terms that are already sitting in the . Those numbers move when structure improves, and they almost never get attributed 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