Your Mega Menu Is the Hardest Thing on Your Site

The most helpful thing on your homepage might be the thing stopping people.

The Short Answer

We measured a university homepage for four different kinds of visitor. For the two with the least room to spare — someone browsing at high magnification, someone with a hand tremor — the single most expensive thing on the page was the mega menu.

Not the forms. Not the text. Not the images. The menu that exists to help them.

Seven times higher. Five times higher. And in both cases it became the hardest single thing they had to do on the entire site.

Somebody Built That Menu on Purpose

Roughly 180 links, tucked behind two buttons in the header.

Nobody did anything wrong. That menu is an act of generosity. Someone sat in a meeting and said people can't find the admissions page, and someone else said let's just put everything one click away, and they were both trying to help. Every link in there is a link somebody asked for.

That is what makes this worth writing about. There is no villain. There is no incompetence. There is a kind decision with a cost that nobody in the room could see.

Why You Cannot See It

Two reasons, and they compound.

It is below the fold. Open that homepage on a laptop and you see a hero image, a headline, three or four buttons. It looks calm. It is calm. The menu only exists once someone opens it, and the person most likely to open it is the person who came for something specific and could not spot it.

Every review of that page — every stakeholder walkthrough, every screenshot in a deck, every "does this look right to you?" — happens above the fold. The expensive part is not in the picture.

It passes. Run any automated accessibility checker over that menu and it comes back clean. The links are labelled. It works from a keyboard. The contrast is fine. There is no violation to report, because nothing about it violates a rule.

The rule set was never designed to ask how much work is this? It asks is this permitted? Those are different questions, and only one of them is about the person.

Compliant and Difficult Are Not Opposites

This is the part most teams have not internalised, and it is not their fault — the tooling taught them otherwise.

A page can satisfy every accessibility requirement ever written and still be exhausting. Nothing in WCAG caps how many choices you may present at once. Nothing flags a menu for being large. A hundred and eighty correctly-labelled links are a hundred and eighty correctly-labelled links.

For a visitor at 400% magnification, that menu arrives a few items at a time through a small window, and they have to hold the shape of it in their head while they scroll. For someone with a tremor, every one of those links is a small target sitting close to other small targets, and choosing means committing to a movement they may have to repeat.

Neither of them is blocked. Both of them are paying. And the tooling has no vocabulary for the difference.

The Number That Matters Is Not the Total

Here is the part that surprised us.

Both of those visitors came out with a low overall score for the page — 0.236 and 0.213, comfortably in the "should be fine" band. If you only looked at the total, you would close the tab and move on.

Their difficulty was not spread evenly. It was concentrated, almost entirely, in one place. The average said fine; the shape said there is a wall two thirds of the way down.

That is the general lesson, and it applies far beyond this one menu: a single score is a sum, and sums hide shapes. A page that is mildly annoying throughout and a page that is effortless until it suddenly is not can produce the same number. They are completely different problems and they need completely different fixes.

Helpfulness Has a Price

The instinct behind a mega menu is real and correct: people cannot find things, so show them more.

But adding an option is never free. It costs a little attention from everyone who sees it, and it costs a lot from anyone who is already spending most of what they have on simply reading the page. The cost lands unevenly, and it lands hardest on the visitors who had the least to begin with.

Which gives you the line to hold on to:

Helpfulness has a price, and it is paid by whoever has the least to spend.

Nobody intends that. It is just what happens when generosity is measured in quantity.

What To Do About It On Monday

None of this requires a tool, a score, or a specialist. Four things, in order of how much they buy you:

Count the choices you hand someone at once. Not links on the page — choices in a single moment, in a single menu, on a single screen. If the number is over about a dozen, you are asking for a decision nobody can make quickly. Most mega menus are ten times that.

Rank them instead of listing them. The five things almost everyone wants, given real visual weight, then everything else behind a "more" that admits it is a long tail. A recommendation is cheaper than a list. A default is cheaper than a choice.

Give search the job the menu is doing. If the menu exists because people cannot find things, the honest fix is a search box that works, not a directory that makes the finding someone else's problem.

Look at your own site below the fold, at 400% zoom. Not a report about it. The actual page. It takes two minutes and it is the single fastest way to see what your reviews have been missing, because you will be looking at it the way the expensive visitor does.

The Uncomfortable Bit

You will not hear about this from your users.

Nobody emails to say your navigation was too much for me today. They open the menu, look at it, feel the small specific tiredness of not knowing where to start, and go somewhere else. In your analytics it is a bounce. It has no cause attached to it, and it never will.

Which means the only way you find out is by going and looking — deliberately, from a position other than your own. That is uncomfortable, because it means the polite silence you have been reading as no news has been telling you something the whole time.

The menu was built to help. It still can. It just needs to be the size of a person's attention rather than the size of your sitemap.

---

The measurements in this piece come from cognitive_effort, which scores the mental work a specific kind of visitor spends on a page and tells you which part of that work breaks them first. The site is left unnamed on purpose — it is a good site, built by people doing their jobs well, and it is not unusual. That is rather the point.