There’s a fascinating conundrum in information architecture: for people to find stuff, it must be organized into choices they recognize. But users understand each choice in a particular context, and they understand the context to the degree they understand individual choices.

The two levels — individual choices and the context where users experience them — influence each other. Setting the right context makes the difference between a place where users can quickly find things and one where they struggle.

Choices (e.g., labels in a navigation menu) represent groups of items. Users will find things more quickly if a) they’re located in logical groups and b) you’ve labeled groups using recognizable terms. There are two challenges here.

First, what might seem logical to one person might baffle another: people have different expectations based on their familiarity with the subject matter. An expert expects things to be in particular groups that use specific labels. Conversely, a new user might need to see more familiar (read: less jargon-y) terms to guide them toward the right groups.

The second challenge is more nuanced — and perhaps harder to tackle. It has to do with context: you always encounter words in some context, and that context changes how you understand what they mean.

For example, consider the verb “save”: its meaning varies depending on what context you’re using it in. “Save” means one thing in a bank, another in a church, and something else still when used in the context of computers. All are valid uses of “save” but refer to different concepts.

The trick is that sets of words — such as the labels that describe options in a navigation menu — create contexts. Consider the following list:

  • Home
  • About Us
  • Vision Care & Products
  • Promotions
  • Patients
  • Blog
  • Contact

You likely recognize these as choices in a website’s navigation bar. But how? It’s because you’ve encountered the labels Home, Blog, and About Us on countless other websites. These three labels provide enough context to let you know this is likely a website.

That said, they don’t provide enough context to tell you what kind of website this is. You’ve probably already guessed this is an optometrist’s website. If you did, it’s due to two other labels in the list: Vision Care & Products and Patients. You don’t see those often on other websites. They change your expectations about what you’ll find under Blog, About Us, and Promotions.

But even as they clarify some things, these two labels also introduce ambiguity. Look at the last option in the list. In most other website contexts, the word Contact is unambiguous: it refers to how you can get in touch with the company. But here, it might have another meaning since optometrists sell a type of product colloquially called contacts.

The term’s location within the navigation hierarchy changes how you understand it. In this case, the ambiguity is mitigated because Contact sits alongside labels like Blog and About Us. If this option referred to the lenses you put in your eyes, you’d expect it to be under Vision Care & Products and not at the top level of the nav.

And, in fact, this site has an option under Vision Care & Products called Contact Lenses. Appending the word Lenses to Contact resolves the ambiguity: with that one move, there can be little doubt left about what you might find there.

The same word — Contact — can have multiple meanings, even within the same navigation structure; whether you understand what it means here depends on its context — what you understand “here” to mean. And you can change the context by increasing the contrast between labels (e.g., appending Lenses to Contact) or changing their location within the hierarchy (e.g., by placing the option under Vision Care & Products.)

This case may sound obvious, but that’s because I picked an example that uses familiar terms. When organizing more complex or esoteric information, you must think carefully about how to group and label things. Research helps to understand the system’s content and users’ mental models and to validate and refine proposed structures.

The challenge is that people experience choices both individually and as part of a set. Context changes the meaning of labels, and the labels in the set change the context — and, therefore, what the labels mean. Getting it right is nuanced work that calls for research, validation, and iteration.

A version of this post first appeared in my newsletter. Subscribe to receive posts like this in your inbox every other Sunday.