Reducing Choices to Reduce Confusion

Ask someone to make their homepage clearer and they’ll usually start rewriting. Better headline, tighter copy. That’s fine, but there’s a cheaper move that works more often, and it doesn’t involve writing a single word. Count the number of things a visitor can click on your page. Then remove half of them.

I mean this almost literally. When I look at audit results for pages that confuse people, the confusion usually isn’t caused by any one bad element. Each element, examined alone, is defensible. There’s a main call to action, and a secondary one, and a navigation bar with seven items, two of which have dropdowns, and a banner about the webinar, and a chat widget, and footer links to everything else. Forty-odd clickable things, each with a reason to exist. The problem is that the visitor doesn’t experience them one at a time. They experience them all at once, as a field of competing claims on their next move. Confusion on a webpage is rarely about not understanding the words. It’s about not knowing what to do, and not knowing what to do is a direct function of how many things you could do.

There’s a useful way to think about this: every choice you present has a cost, paid not by you but by the visitor, in attention. The costs are small individually. Scanning a menu item takes a fraction of a second; deciding to ignore the banner takes another. But a first-time visitor arrives with only a few seconds of willing attention, and the taxes add up fast. When the total exceeds the budget, the visitor doesn’t carefully pick the best option. They do what people do in any overwhelming store: leave without buying. The tragedy is that they’d often have been happy with what you sell. They just couldn’t afford the cost of figuring out where to start.

Why do pages accumulate choices? Never because anyone decides “let’s have forty links.” It happens the way sediment happens. Each team adds its own door: sales wants the demo button, marketing wants the webinar banner, support wants the chat bubble, the founder wants the blog in the nav. Every addition passes review, because every addition is individually reasonable and nobody is charged with defending the whole. A page has many parents but no editor. And subtraction has a structural disadvantage: adding a link visibly serves someone inside the company, while removing one serves only the visitor, who isn’t in the meeting.

The fix requires deciding something teams hate deciding: what is the one thing you most want a first-time visitor to do? Not the three things. One. For most product sites it’s start a trial, see a demo, or, in the case of my own site, run a scan. Once you’ve named it, the editing rule follows: that action gets visual dominance, one secondary action is permitted for people who aren’t ready, and everything else must justify its existence against the question “does this help a stranger take the main action, or does it merely help some department feel represented?” Applied honestly, that question clears a page the way a good editor clears a manuscript.

The pushback is always the same: but some visitors need the pricing page, some need careers, some need documentation. True, and this is where nuance matters, because reducing choices doesn’t mean amputating capability. It means managing prominence. The visitor who needs your careers page is on a mission; they will find a small footer link, because motivated people search. The visitor you lose to clutter is the opposite kind: unmotivated, evaluating, one ambiguous second away from the back button. Prominence is the scarce resource. Spend it on the undecided, because the decided will spend their own.

There’s a version of this discipline for words too, not just links. A page that makes three claims lands harder than a page that makes nine, for the same budget-arithmetic reason. But I focus on clickable elements because they’re countable, and countable things actually get fixed. You can’t easily measure “how foggy is our messaging,” but anyone can count buttons before and after. When my audit tool evaluates conversion and clarity together, an overloaded action landscape is among the most common findings, and the prescribed remedy is usually demotion rather than deletion: keep the capability, move it down, shrink it, fold it into a menu, until one path is unmistakably the path.

The deeper point is that a homepage is not an inventory of what your company offers. It’s an answer to a stranger’s question: what should I do next? A good answer is singular. If you find yourself unable to remove anything because every element has a constituency, that’s worth noticing, it means your page is organized around your org chart instead of your visitor. The visitor doesn’t know your org chart. They know they have five seconds and forty doors, and mathematics is against you.