How to A/B test a landing page with low traffic

Most landing pages don't have the traffic for reliable A/B tests on small changes. Here's how to test anyway: bigger changes, better metrics, longer tests, and faster ways to decide what to test.

The Klar team ·

An A/B test is the most reliable way to know whether a change improved your landing page. It's also the slowest, and on a page with a few thousand visitors a month, a test on a small change can take months or never reach a clear result.

That doesn't mean you can't test. It means you have to test differently.

Why low traffic is a problem

The number of visitors a test needs depends mostly on two things: your current conversion rate, and the size of the change you want to detect (the minimum detectable effect).

As a rough example, a page converting at 3% needs about 14,000 visitors per version to reliably detect a 20% relative lift (from 3% to 3.6%) at standard settings. To detect a 10% lift, it needs roughly four times as many. Try your own numbers in our A/B test sample size calculator.

If your page gets 5,000 visitors a week, that first test takes about six weeks. The second, on a 10% lift, takes about five months. That's the real constraint: not whether you can test, but how many tests you can afford to run in a year.

1. Test bigger changes

Small changes, like a button colour or a word in the subline, usually have small effects, and small effects need huge samples. Big changes are more likely to move the conversion rate enough to measure:

  • A different value proposition or headline angle
  • A different offer: free trial vs. demo, free plan vs. trial
  • A different page structure: short vs. long, proof first vs. product first
  • Adding or removing a step, like a form field or a pricing section

These are also the changes most likely to matter for the business.

2. Measure something closer to the change

If your primary conversion is rare (say, booked demos), measure a step that happens more often and sits closer to what you changed:

  • Clicks on the main call to action
  • Starts of the sign-up form
  • Scroll to the pricing section

These micro-conversions happen more often, so tests finish sooner. The trade-off: a lift in clicks doesn't always mean a lift in sign-ups, so check the final conversion too, even if it can't reach significance on its own.

3. Run for whole weeks, and decide the length up front

Use a sample size calculator before you start, and commit to the length. Run whole weeks so weekday and weekend visitors are both counted. Don't stop early because one version looks ahead: with small samples, early leads are often noise, and stopping at the first good-looking moment makes false positives much more likely.

4. Accept a bit more uncertainty, knowingly

For low-risk changes, you can lower the confidence level (for example, 90% instead of 95%). The test needs fewer visitors, at the cost of a higher chance of a false positive. That's a reasonable trade when the change is cheap to reverse; it's not when it touches pricing or the core offer.

5. Test where the traffic is

If one page gets most of your traffic, test there. Changes to a shared element, like the main headline pattern or the sign-up form, can sometimes be tested across several pages at once.

6. Spend less time testing ideas that won't win

With only a few tests a year, the biggest gain is in choosing what to test. Use faster methods to filter ideas first:

  • Talk to customers about what nearly stopped them from signing up.
  • Run a five-second test on headline options.
  • Watch session recordings of visitors who left at the pricing section.
  • Ask a simulated panel. Klar's Copy Lab puts up to 5 versions of your headline, subline and call to action in front of the same five buyer personas that reviewed your page, and shows which version each one picks and why. It takes minutes and doesn't use any traffic.

None of these replace the test. They help make sure each slow test is spent on an idea with a real chance of winning.

When not to A/B test at all

Sometimes the right answer is to skip the test:

  • The page is clearly broken. If visitors can't tell what you sell, fix it. You don't need a test to confirm that clarity is better than confusion.
  • The change is required anyway, for legal, brand or product reasons.
  • The traffic will never be enough. A page with a few hundred visitors a month is better served by customer interviews and careful judgement than by a test that can't conclude.

Summary

With low traffic, test fewer, bigger ideas; measure the step closest to the change; commit to a test length up front; and use fast methods to decide what's worth testing. Our sample size calculator tells you how long each test will take before you start.