The five-second test I run on every homepage

Before I redesign any homepage, I run the same cheap test. It takes five minutes, costs nothing, and it predicts most of what a full audit would find.

Here's the test: show the homepage to someone for five seconds - a timer, honest, five seconds. Hide it. Then ask three questions:

1. What does this company sell? 2. Who is it for? 3. What would you click next?

That's it. The whole test.

Why it works

Five seconds is roughly the attention a stranger gives your site before deciding to stay or leave. If your core message doesn't land inside that window, it doesn't land at all - the visitor is already gone, and all the beautiful content below the fold was written for nobody.

The third question is the sneaky important one. "What would you click next?" tests whether your page has a clear next step. If they say "I don't know, maybe the menu?" - your call to action is decoration, not design.

What the results usually look like

• They can't name what you sell → headline problem. Say the thing, plainly.
• They can't tell who it's for → you're writing for everyone, which reads as no one.
• No idea what to click → you have four CTAs, which means you have zero.

Run it on your own site

Honestly, don't run it on your own site - you already know the answers, which ruins the test. Run it on a friend, a customer, anyone who isn't invested. The gap between what you think the page says and what they hear is the actual work.

If you run it and don't like the answers, that's a fixable problem. It's most of what I do.