How to test a landing page before launch

The usual plan is to launch, buy some traffic and watch the numbers. That costs money, takes weeks to reach a clear answer, and still does not say why people left. Most of the problems that sink a landing page are visible before launch, if someone who has never seen it looks at it with a reason to leave.

• A technical pass you can do in an hour, with nobody else involved.
• Five questions every stranger should be able to answer from the first screen.
• Ways to get outside eyes on the page, from free to paid, and what each one misses.
• How to retest after a fix, so you know the fix worked.

A conversion rate needs real visitors arriving from real channels, so no pre-launch test can give you one. A tool that quotes a predicted rate for a page nobody has visited is claiming more than it can know.

What you can learn is whether the page does its basic job: a stranger understands the offer, believes it enough to keep reading, and knows what the button will do. Those failures are common, they are visible on the page itself, and they are cheap to fix before money is spent sending people to it.

Open the page on a phone, on mobile data rather than office wifi. Note how long the first screen takes to appear and whether anything jumps around as it loads. Lighthouse, built into Chrome's developer tools, gives a quick read on speed and basic accessibility.

Click every link and button. Submit the form with an address you can check, and confirm that the thank-you page and any confirmation email arrive. Then check that your analytics or ad platform recorded the sign-up, because a page that converts without tracking looks exactly like a page that does not convert.

Check the share preview too. Paste the address into a chat app or a social post draft and see which title, description and image appear. Plenty of launch posts go out with a blank or placeholder preview.

What is it? From the headline and the line under it, a stranger should be able to name the kind of product. If your headline could sit on a competitor's page unchanged, it is not doing this job.

Who is it for? Visitors decide fast whether a page is meant for someone like them. Name the audience or the situation, not "teams" or "everyone".

Why believe it? Every claim needs something behind it: a screenshot of the real product, a specific example, a customer you actually have, or a number you can defend.

What does it cost, and what is the catch? Missing pricing reads as expensive, or as a sales call. If you cannot show a price, say how pricing works and what the first step costs.

What happens when I click? "Get started" says nothing. Say what the button does and whether a card is required.

You already know what the product does, so you cannot see the page cold. Colleagues and friends cannot either, and they are polite. The useful reader matches your audience and owes you nothing.

A five second test shows the first screen briefly and asks what people remember, which checks the headline and the visual hierarchy. Watching a few people from your audience try to make sense of the page, while you stay quiet, catches the rest. Both use real people and take some organizing.

Simulated visitors are the fastest option. In a Mimiq page test, 25 simulated people matched to your audience land on the live page, decide whether they would sign up, stay, skim or leave, and say why. The objections come back ranked by how many people raised them. They are simulated, so treat them as a list of likely problems to check, not as measured behavior.

Read a full sample report How a five second test works

After a change, show the new version to the same readers. If different people see each version, you cannot tell whether the page improved or the readers changed. In Mimiq, run the new version on the same saved audience, and the comparison shows who changed their mind and whether the difference is likely to be chance.

Read the reasons more than the call. On 1,000 real headline A/B tests held out from its development, Mimiq's forecast picked the headline more readers clicked 76% of the time; the best rule of thumb, picking the headline written first, got 61%. Mimiq's forecast has held up on headlines. In pre-registered tests on emails, text messages and ads, it was no better than chance at ranking small wording changes, and on real web page tests it did not beat the simplest guess. So on a whole page, the useful part of a comparison is which objections went away.

Change one thing at a time where you can. If the headline, the hero image and the pricing all change at once, a better result does not tell you which change helped.

Once traffic arrives, real behavior beats any pre-launch test. Watch where visitors drop off in your analytics, look at recordings of sessions that ended early, and, with enough traffic, run a live A/B test on the questions the pre-launch checks could not settle.

Keep testing each new version before it goes live. It decides which changes deserve a slot in a live test, since each live test costs weeks of traffic.

A/B testing with low traffic