Mtools

Content

Marketing

Your website menu follows the org chart, and buyers don't know your org chart

Visitors arrive with a problem and no idea which department of your company it belongs to. See how to organize the menu around buyers' questions and test it with five people.

A three-panel folded map in frosted glass 3D, with a red location pin glowing inside, on a black background

When your website information architecture copies the org chart, with one page per department and every product under its internal name, visitors have to guess which box their problem lives in. Buyers arrive with a question and don't know the names of your departments. The fix is to organize the menu around buyers' questions and test the structure with five people in a single afternoon.

World Usability Day, held on the second Thursday of November since 2005, falls on November 12 in 2026. It's a good excuse to open your company's site through the eyes of someone who has never seen the org chart.

The signs of a site organized from the inside

A site follows the org chart when the menu speaks the company's language. The most common signs:

  • menu items named after departments, such as “Corporate division” or “Services unit”;
  • products under their internal name or the code name of the project that created them;
  • pages that exist because each department asked for one, with copy that repeats;
  • a hidden contact option, only in the footer or on a generic page;
  • the same answer scattered across pages from different departments, each holding a piece.

The result shows up in the report as visits that don't turn into contacts. The blame usually lands on design or page speed, when the problem is the website information architecture.

The question the visitor brings

Every visitor arrives with a question, and the menu has to show where the answer is. In complex sales, people evaluating a purchase tend to arrive with questions like these:

  • do you solve my problem?
  • does it work for a company in my industry and of my size?
  • who already uses it and what results did they get?
  • how does implementation work?
  • how much does it cost, or how do I buy?
  • how do I talk to someone?

Each question becomes a path in the menu: solutions by problem, industries served, case studies, how it works, how to buy and a contact option visible on every page. The purchase involves several people, so the same menu has to serve the people who decide the purchase and the people who will use the product.

The real questions are in the sales team's notes, in the messages from the contact form and in the searches that bring visits from Google.

How to test your website information architecture in an afternoon

Two simple tests show whether buyers find what they're looking for: card sorting shapes the menu, and task testing checks whether it works. Both fit in an afternoon, with five people similar to your customers, ideally matching the profile of a persona built from real customers. Leave your own team out, because they know the org chart.

Card sorting: buyers group the cards

In card sorting, each participant places cards with the site's content into groups. In the open version, they create and name their own groups; in the closed version, they fit the cards into predefined categories, as Nielsen Norman Group explains.

The same source recommends at least 15 participants for a card sort. With five in an afternoon, you get the first clues for a draft menu and complete the sample later with an online tool.

Task testing: five people are enough

For task testing, five people do the job. By Jakob Nielsen's math, tests with five users find about 85% of usability problems, as long as they use the site in similar ways, and several small tests beat one big one. Step by step:

  1. Write five to eight tasks phrased as a buyer's question, without the word that's in the menu (“you want to know whether the solution works for a logistics company”).
  2. With permission to record, show the site or the draft menu and ask the person to point to where they'd click, without help.
  3. Note the first click and whether they reached the right page.
  4. Add up the successes per task. Any task where two or more people got it wrong is the priority.
  5. Adjust the menu and repeat with five other people.

When we built Orbenk's corporate website, information architecture was a stage of its own, before the wireframes, with the pages most relevant to the audience highlighted. The menus were defined together with the copy, so the client could approve everything at once. That's how our website development projects start: before any screen, the audience and what the site needs to convert.

Talk to usWant help planning this at your company?Tell us what you need. We'll reply by email or WhatsApp and, if it makes sense, put together a quote.Contact us

What to measure after the change

The change worked if more visitors reach the commercial pages and more of them ask for a contact. Compare the same period before and after:

  • the success rate in the task test, repeated with the new structure;
  • visits to case studies, how to buy and contact;
  • the site's contact conversion rate;
  • the questions sales still hears that the site should answer.

If visits to commercial pages went up and contacts didn't, the problem moved. Check what to change in the form before asking for more budget and see where the company loses leads after the click.

How to win over the departments that lost their spot in the menu

The argument that wins departments over is the test recording. Watching five people fail to find an offer carries more weight than any opinion about the menu.

In the conversation with each department, three agreements help:

  • the department's content stays on the site, now inside the question it answers;
  • the org chart gets its place on the company page, for people who want to know the company;
  • a review date, with the numbers from the previous section on the table.

Frequently asked questions

What is website information architecture?

The way pages are organized, named and linked to each other, from the menu to internal links. It decides whether visitors find the answer they're looking for without guessing the path.

How many people do I need to test my website's usability?

For a task test, five people similar to your customers already reveal most of the problems. For card sorting, Nielsen Norman Group recommends at least 15.

What's the difference between card sorting and task testing?

Card sorting helps you build the menu, because it shows how buyers group and name the content. Task testing checks whether the finished menu works, with real questions and the path each person takes.

Share

shall we createsomething together?

related content