Your users can't find what they need. The clues are already in your data.

They leave traces. Users leave evidence everywhere they go: in your analytics, in your search logs, in the recordings of their sessions. The problem is not usually a lack of data. It is knowing what you are looking at.

Before you can fix a findability problem, you need to be clear about what the problem actually is. Not a vague sense that "the navigation feels off," but a specific, defensible statement you can act on. That definition is what separates a redesign that solves something from a redesign that just shuffles the furniture.

Your analytics platform is the obvious starting point, but most people look at the wrong things. Bounce rate is not especially useful on its own. What you want to watch for is the Google loop: users arriving from search, visiting a single page, returning to Google, and coming back again. They are using a search engine to navigate your site because your navigation is not doing the job.

Other signals worth tracking: high drop-off rates within flows that should be straightforward, and a disproportionate volume of contact requests for information that is already on the site. Both suggest users cannot find what they came for.

If your platform gives you access to search query data, this is one of the richest diagnostic sources available. Look for users searching for the labels in your own navigation. That is a strong indicator that your labels are not communicating what they contain. Look also for searches that suggest users are trying to locate something within a page rather than across the site, which points to a different problem entirely: content structure, not navigation.

Tools like Hotjar or Crazyegg show you what hesitation looks like. Repeated scrolling up and down a page, a pause before clicking anything, moving to the footer to navigate because the header is not working. These are not abstract data points; they are people struggling in real time.

This is where you get the clearest signal, and the most context. Users will show you exactly where they get stuck, and if you are present you can ask them why. One caveat: think carefully about who you are recruiting. A participant who does not regularly use the device type you are testing on will encounter problems that belong to your recruitment process, not your design.

A well-timed exit survey asks users a single question at the moment they are leaving: did you find what you were looking for? If not, what were you looking for?

That second question is the important one. A yes/no answer gives you a proportion. The open text gives you the actual gaps, in users' own words, which is far more useful than any percentage score.

Keep exit surveys short and specific. One or two questions at most. The moment they become a form, users stop filling them in, and you lose the signal entirely.

High volumes of support requests for information that exists on the site are a findability problem wearing a customer service hat. If users cannot locate something, they ask. Every repeated question is a gap in your navigation or content structure, and quite possibly both.

Once you have gathered enough signal, the goal is a clear problem statement. Not "navigation is confusing," but something specific enough to guide what comes next.

For example: mobile users are abandoning product pages at a higher rate than desktop users, and session recordings show consistent hesitation around the category structure. That is something you can investigate further. That is something you can test against.

Whether you formalise this in a document or keep it as a working hypothesis depends entirely on your context. What matters is that it is precise enough to be useful, and honest enough to rule out the things that are not actually the problem.

It is tempting to move straight to solutions. A new navigation structure, a revised taxonomy, a different search configuration. But without a clear problem definition, you are making changes based on assumption rather than evidence.

The data is already there. Users have already told you where things are breaking down. The work is in reading what they left behind, and using it to define something worth solving.

Want to know whether your site has a findability problem? The Murmuration diagnostic is designed to find out.