Draft: shown until launch for review, and not indexed.
How to test a homepage before your customers see it
A ten-minute routine for finding what customers will notice, question and be put off by on a new homepage, before you ship it and before you pay to send traffic to it.
Key takeaways
- Test the page with people who resemble your customers, not with your team.
- Watch what they look at first, then read what they say about it: their own words show what puts them off.
- Ask why about one reaction at a time, then change the page and show them again.
Most homepages are reviewed by the people who wrote them. They know what the product does, so they read the page the way it was meant, not the way it reads. The first honest review usually arrives after launch, from customers who leave.
You can get that review earlier. Here is a routine that takes about ten minutes, with POVS or with a few real people.
Who should look at it?
Pick three people who would actually buy, and make them different from each other. A customer who already pays for something similar, a skeptic, and someone with no context at all is a good start.
- A customer like your best ones. Describe them in a sentence: what they already use, and what would make them switch.
- A skeptic. Someone who needs proof before believing a claim.
- A first impression. No brief, no context: does the page make sense on its own?
What should you watch for?
Not their opinion first. Watch where they look and in what order. The phrase someone selects is usually the one that changed how they read the page.
| Signal | What it usually means |
|---|---|
| They read the headline twice | It is clever, but not clear |
| They go straight to the price | The value is not landing above it |
| They hesitate before the call to action | Something earlier raised a doubt |
How do you ask why without leading them?
Ask about one reaction at a time, and start with why? Then ask what would change their mind. Do not argue: if you want a new opinion, change the page and show them again.
I understand what this does, but I still don’t know why I would choose it.
Questions that work
- Why did you stop there?
- What did you expect to see here?
- What would make you trust this?
What do you do with the answers?
Fix the one thing that came up most, then run it again. Two rounds before launch beat ten opinions after it.
Questions
How many people do I need to test a homepage?
Three different people find most of the problems that matter. Make them different from each other, not more of the same.
Can I test a page before it is live?
Yes. With POVS you can test anything on your screen: a Figma frame, a build on localhost, or a screenshot.