Everything is in our navigation, so why can't people find anything?

Nobody reads your navigation. They scan it for about two seconds, looking for a word that matches the thing they came to do. If that word is not there, the menu has failed, even if the page they wanted is sitting three items away under a label they did not recognise.

This is why adding a link almost never fixes a findability complaint. Every item you add lowers the odds that any single item gets noticed. A menu with five items gives each one real attention. A menu with eleven gives each one a glance. The eleven-item menu contains more and communicates less.

So the honest version of the question is not “what is missing?” It is “which four words are wrong, and which two items are stealing attention from the ones that matter?”

Menus get built in meetings. Each team defends its territory, and the labels that survive are the ones that make sense inside the building. “Solutions”, “Platform”, “Resources” and “Company” are four words that tell a visitor nothing about whether you do the thing they need.

A visitor arrives thinking “how much does this cost”, “does it work with our system”, “is there someone I can talk to”. None of those words appear in that menu. They appear two clicks down, inside a page that was named after a department.

Across the redesigns we run, the pattern repeats: the nav item the internal team argues hardest about is usually the one visitors never click. The argument is about who owns it, not about who needs it.

This costs nothing but an afternoon and five conversations. Do the steps in order, because each one narrows what the next one has to decide.

Do not run step one as a workshop. A workshop produces the words your team uses. You want the words your visitors use, which are almost always shorter and blunter.

1. Write down the five things visitors actually come to do, in their words. Pull them from your site search queries, the first two minutes of sales calls, and the subject lines of support tickets.
2. For each task, ask someone outside your company to point at the nav item they would click. Do not let them click. Note where they hesitate, not only where they are wrong. A two second pause on the right answer still means the label is weak.
3. Mark every item nobody chose for any of the five tasks. These are candidates for the footer or a single landing page, not for deletion.
4. Rename items using the words people said out loud. If two items compete for the same task, one of them is the wrong word. Pick the one people reached for first.
5. Recount the top level. Over roughly seven items and you are no longer listing, you are ranking. Decide what loses.

Each of these is a decision that was avoided rather than made. Naming yours makes the fix obvious.

• Nav as org chart. Labels like "Solutions" and "Platform" only parse if you already work there. A visitor cannot tell which one holds the pricing.
• The mega-menu as a truce. Every team gets a column, so nothing is prioritised. It is a sitemap wearing a costume, and it hides the top three links inside forty.
• Duplicate doors. The same page reachable under three labels. Attention splits three ways and your analytics become unreadable, because no single path ever looks important enough to protect.
• A sticky mobile nav that eats a third of the screen while still hiding the one link people wanted. The bar is always visible. The answer is still two taps away.
• Search as an excuse. Adding site search so nobody has to decide what the menu says. Search then fills with the queries that prove the menu is broken, and nobody reads them.

The fear is always the same: traffic to that page will collapse. In our projects it does drop, often by a lot, and that is usually the point. If a page was getting clicks only because it sat in the top navigation, those clicks were borrowed from the pages that convert.

Here is the shape of it. A company had “Integrations” in the top nav because the partnerships team asked for it. In our five-task test, nobody pointed at it for any task, including “does it work with our system” (they pointed at “Product”). We moved it to the footer and linked it from the product page where the question actually gets asked. Direct visits to the integrations page fell. Clicks on “Pricing”, now one of five items instead of one of nine, rose. The complaint that started the project, that people were emailing sales to ask where pricing was, stopped.

The counterexample matters too. If people do point at an item during the test and still cannot find the page behind it, the problem is not the menu. It is that the page does not exist or answers a different question than its label promises. Renaming will not save it.

Renaming four items and moving two to the footer is a content edit. On most CMS setups it is an afternoon of work for someone on your team, and no designer needs to be involved. If a studio responds to “people can’t find our pricing” with a proposal for a full rebuild, ask them what the five-task test showed first.

There is one case where the menu genuinely cannot be fixed by editing it: when the tasks people arrive with have no page to point at. If your five tasks are “see pricing”, “check it works with our stack”, “talk to someone”, “see if a company like mine uses this” and “understand what it actually does”, and three of those have no home, no label will help. You are not missing words, you are missing pages.

Tell the difference like this. Write each of the five tasks next to the URL that answers it. If every task has a URL, it is a naming job you can do yourself. If two or more tasks have no URL, or share one bloated page that answers all of them badly, the page structure is the problem and that is redesign territory.

Give it four weeks before you judge anything. Less than that and you are reading weekly traffic noise. Longer than that and you have lost the ability to attribute the change to the edit.

Track three things, in this order of importance. First, menu click share per item: what percentage of nav clicks each item gets. A healthy top level has no item under about 3 percent. Anything below that is decoration. Second, tasks completed without using site search. If your search box gets fewer queries for the five words you renamed around, the labels are working. Third, the complaint itself. Go back to the sales calls and support tickets you pulled the tasks from, and count how often “I couldn’t find” still appears.

The last one is the real measure. A menu is not there to be complete. It is there to make five things easy to find, and everything else findable by someone who is already looking.