Why more choices means fewer calls · Shoebill Design Studio

Every option you add to a page is a decision the visitor has to make, and a decision takes time. When somebody arrives with a problem and finds four different ways to act on it, some of them resolve the ambiguity by doing nothing and coming back later. Later rarely happens.

This is the opposite of how options usually get added. Each one gets added by somebody solving a real problem: the chat widget for people who don’t like forms, the popup for the offer, the second button for the other service. Individually reasonable. Collectively a maze.

Why choice slows people down

The response-time research goes back to the 1950s: the time it takes somebody to choose grows with the number of options. Hick’s law, in the original work on reaction time, and it’s about milliseconds in a lab rather than website conversions.

But the mechanism transfers, and you can watch it happen: a page with one obvious action gets acted on, and a page with four gets scanned, considered and left. The delay isn’t the milliseconds. It’s that a small ambiguity gives somebody an excuse to defer, and deferring is always the easiest option available.

Every equal-weight option you add makes doing nothing relatively more attractive.

What overload looks like on a real page

The common set, all at once, all on a phone:

A sticky header with a phone number and a button
A hero with two buttons of equal weight
A chat bubble in the bottom right
A popup after five seconds
A cookie banner across the bottom
A navigation menu with eleven items

None of those is indefensible on its own. Together, the visible screen is asking the visitor to make five decisions before they’ve read a sentence about whether you can help them.

The version of this that costs the most isn’t visual clutter. It’s when two things ask for the same click. A chat bubble and a quote button are both “get in touch,” so the visitor has to work out which one gets them a real answer faster, and that question has no obvious answer.

The shape that works

One primary action per screen. One quieter secondary. A phone number that’s visible and tappable rather than a third button.

That gives three ways to act, one of which is obviously the main one, which is not the same as three competing options. The visitor who wants to call, calls. The visitor who wants to think, uses the form. Nobody has to decide which one you’d prefer.

For navigation: group by what somebody is trying to do rather than by your internal structure, and cut anything nobody clicks. A nav that lists every service is a nav that asks the visitor to categorize their own problem using your categories, and they don’t know your categories.

What you can check in the next thirty seconds

Take a screenshot of your busiest page on a phone. Then circle everything on that screenshot that’s asking for a click.

Count the circles. If there’s more than one plus your phone number, you’re asking the visitor to choose.

Then ask a harder question about each circle: what happens if I remove this? If the honest answer is “nothing,” it isn’t helping. If the answer is “somebody would complain,” find out whether that somebody is a customer or a colleague.

The one that’s hardest to cut

Chat widgets, usually, because they feel like service.

The test is the transcripts. Open last month’s and read them. If they’re real conversations that turned into work, keep it and make your button quieter. If they’re mostly nobody, or bots, you’re paying load time on every page for a widget that competes with the action you actually want, and the load time is not small.

That’s a decision about evidence rather than taste, which makes it one of the easier ones to settle.

When to bring somebody in

Removing things is free and you can start today. Screenshot, circle, cut what nobody uses.

Where it becomes a build is when the page has nowhere sensible to put a single clear action, or when the competing elements are baked into the template so removing one breaks the layout. Then it’s a structural question: what is this page for, what’s the one thing it should ask, and where does that thing live on every template.

That’s part of how we build websites. The rule we hold to is one action per screen, in the same place on every page, named for what the visitor gets.

Common questions

So should I remove the chat widget?

Only if it is not producing real conversations. Check the transcripts from the last month. If they are mostly nobody, it is costing you load time and competing with your main action for nothing.

What about giving people a choice of phone or form?

That is a good pair, because they suit different people at different moments. It works when one of them is clearly the main action and the other is available, rather than both being equal-weight buttons side by side.

Does this apply to navigation?

Yes, and it is where most sites overload first. A nav with eleven items asks the visitor to categorize themselves before they have read anything. Group them, or cut the ones nobody clicks.

Is a popup always bad?

It is an interruption, so it needs to earn the interruption. One that appears immediately, before anybody has read anything, mostly teaches people to dismiss things on your site. Timing and what it offers decide whether it is worth it.

How do I know what to cut?

Look at what gets clicked. Anything that has not been used in a month is not serving visitors, it is serving whoever asked for it, and those are different things.

Consumer Psychology

Why the first thing somebody reads about you decides the rest

People anchor on the first piece of evidence they see and read everything after it in that light. What that means for where your proof sits on the page.

4 min readAug 29, 2026