More action from the visitors you already have.

It’s the process of getting more of your visitors to do the thing you want — join, buy, donate or get in touch. We find where people give up, work out why, test a fix with real visitors and keep only the changes that are proven to help.

As a rough guide, a page with a few thousand visitors a month and a steady flow of conversions is enough to run reliable A/B tests. If you have less, we’re upfront about it: we focus on research and user testing instead, and fix the obvious problems without waiting for a test to tell us.

Usually two to four weeks. A test needs enough visitors to give a trustworthy answer, and should run for full weeks so weekday and weekend behaviour are both counted. We never stop a test early just because it looks good.

That’s still useful. A losing test tells you what not to change, so you avoid rolling out something that would have quietly cost you sign-ups. Every result, win or lose, teaches us more about your visitors and sharpens the next test.

No, not noticeably. We use lightweight testing tools set up to load quickly and avoid the page “flicker” you sometimes see. We check speed before and after, and remove test code as soon as a test ends.

Yes, a lot. More sign-ups, renewals and donations from the visitors you already have is often the most cost-effective growth there is, especially when budgets are tight.