Menu

IA

Label Testing

A practical information architecture method for checking whether the words you use in navigation, categories, and actions make sense to users.

How to use label testing to improve terminology, reduce confusion, and make navigation and content easier for users to understand and choose confidently.

4 min read

What it is

Label testing is a UX method used to evaluate whether the words used in , buttons, categories, and content make sense to users.

It focuses on , meaning, and interpretation rather than .

Users are shown labels in or isolation and asked what they think they mean, where they expect them to , or which option they would choose for a task.

This method is closely tied to and , as poor is one of the most common causes of confusion.

The goal is to ensure that terminology is clear, intuitive, and aligned with how users think.

If users have to stop and interpret the label, the wording is already creating friction.

When to use it

Use this method when language and matter.

It is most useful when:

You are designing or refining navigation
Users are struggling to find content
You are testing category names or menu items
You are improving clarity and usability
You are working with complex or unfamiliar terminology

It is less useful when:

You are testing full interaction or journeys
Language is not a key issue
You need deep behavioural insight
Label testing is often used alongside card sorting, tree testing, and findability testing.

Key takeaway

Use label testing when the main problem is not where things are placed, but whether the words used to describe them make sense to users.

How to run it

Set up properly

Be clear on which labels you are testing, what you expect people to understand by them, and what counts as a correct interpretation. Testing labels without a defined meaning produces opinions rather than findings.

Present labels in a realistic wherever you can. A word in a list behaves differently from the same word sitting in a bar next to five others.

Run the method

Label testing is about interpretation. It is one of the highest-return things you can test, because a label is the smallest unit of and the cheapest thing to change.

  1. Show the label or option without explanation or surrounding help text.
  2. Ask what people think it means, in their own words, before asking anything else.
  3. Ask where they expect it to , and what they would expect to find when they arrive.
  4. Give a task and ask which label they would choose. Stated understanding and actual choice frequently disagree.
  5. Capture reasoning and together. A correct choice made hesitantly is a label that will fail under real conditions.

Avoid leading or explaining. The moment you clarify what a label means, you have taught the participant the thing you were trying to measure.

Capture and make sense of it

The value comes from understanding interpretation. Look across to identify:

  • Labels consistently understood as something other than intended
  • Ambiguity, where different people read the same word differently
  • Gaps between your terminology and the words people volunteered
  • Labels chosen correctly but with low

Use this to refine wording and structure. Where a label keeps failing, the problem is often the category behind it rather than the word in front.

What to look for

Focus on:

Clarity: whether labels are understood
Interpretation: what users think they mean
Hesitation: correct choices made slowly, which fail under real conditions
Expectation: what people think sits behind the label
Consistency: whether understanding aligns across users

Where it goes wrong

Most issues come from:

If users have to think, the label is wrong.

Showing a label with no surrounding context, where people guess
Phrasing the question so it contains the answer
Defending internal terminology because it is technically accurate
Recording the choice without the confidence behind it
Changing a word without testing whether the replacement is any clearer

What you get from it

Done properly, this method gives you:

Words your audience actually understands, not ones you have agreed internally
Labels chosen correctly but hesitantly, which will fail under real conditions
The vocabulary people volunteer, which feeds navigation and search
The cheapest structural improvement available, since a label is one string

Key takeaway

It helps you say things in a way users actually understand.

Get in touch

If this sounds like something you need, we can help you simplify your language so users instantly understand where to go.

No guesswork. No assumptions. Just clarity that works.

FAQ

Common questions

A few practical answers to the questions that usually come up around this method.

What is label testing in UX?

It is a method used to evaluate whether labels and terminology make sense to users.

When should you use label testing?

Use it when refining navigation, categories, or content language.

How is it different from card sorting?

Card sorting focuses on grouping content, while label testing focuses on naming.

What types of labels can be tested?

Navigation items, buttons, categories, and content labels.

Does label testing improve UX?

Yes. It ensures users can understand and navigate with confidence.

Quick take

If users don’t understand your labels, they won’t find anything. Use label testing to fix that.

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