A client emails me a link to their site. Before I do anything thoughtful, I do something stupid: I open it on my phone, look at it for the time it takes to put a kettle on, and close it. Then I write down what I remember. That's the homepage exercise. Almost everything I'm going to recommend later lives in what I forgot.
Seven seconds is the number people quote. I don't think the number matters. What matters is that nobody is reading. They're glancing, deciding whether you're worth a second glance, and then moving on. Your homepage doesn't need to be good. It needs to survive a glance.
The three questions a stranger is actually asking
When I do this exercise on a client's site, I'm not grading their design. I'm trying to answer three things on their behalf:
- Is this for me? Does this look like a business that solves a problem I actually have, in a place I actually live?
- Are these people serious? Does this feel like a real operation, or a half-finished template?
- Where do I click? Is the next move obvious without scrolling, reading, or guessing?
If a stranger can't answer all three on instinct, they don't bounce because you're bad. They bounce because they ran out of patience for figuring out whether you're good.
A small note on speed: every extra second your site takes to load is a small percentage of mobile visitors walking out. You don't need a perfect Lighthouse score. You just need to not be slow.
The three things I see missing, almost every time
A headline that talks about the visitor, not the business
"Welcome to Smith Family Plumbing — serving Austin since 2005" is a headline written for the family, not for the person whose sink is leaking. Try saying out loud what the visitor actually needs to hear. Often it's something embarrassingly direct: "Plumbers in Austin. Same-day calls answered." The flat version sounds less like marketing, which is exactly why it works.
A next step that isn't competing with three other next steps
Most homepages have a primary button, a secondary button, a phone number in the header, a chat widget in the corner, and a newsletter pop-up. From the visitor's side, that's five things asking for attention. Pick the one action that matters most. Make it the loudest thing on screen. Demote everything else.
A mobile version that wasn't an afterthought
More than half your visitors are on a phone, often on a slow connection, often standing up. If your menu requires precise tapping, if your hero image takes a beat to render, if your text shrinks below 14px, none of the rest of this matters. Open your site on a friend's phone (not yours) and watch them try to use it. The flinches are the bugs.
What I'd ask you to do this afternoon
Open your homepage on your phone, in a coffee shop, somewhere with patchy Wi-Fi. Count to seven. Close the tab. Write down everything you remember. If the list doesn't include what you do, who it's for, and how to start, you have a homepage problem worth fixing.
You probably don't need a redesign. You probably need a sharper headline, one obvious button, and a faster load. That's a couple of evenings of work, not a quarter-long project.
If you'd rather not stare at your own site for a couple of evenings, send it over. I'll do the seven-second exercise on it and write back what I noticed. Free homepage read, no follow-up sales call.
