From install to your first test in about fifteen minutes.

Copy one line of JavaScript and paste it before the closing </head> tag on your site. That's the entire install. It works with static HTML, WordPress, Webflow, Squarespace, Shopify, or anything else that lets you add a script tag.

The snippet is under 8KB, loads asynchronously, and is built to stay out of the way. Once it's live, SplitPea will detect the connection and you're ready to create your first experiment.

Every experiment starts with a brief. You give it a name, write down what you think will happen (your hypothesis), pick the page you want to test, and choose a conversion goal.

The hypothesis part is optional but worth doing. Writing down "I think a shorter headline will get more signups" takes ten seconds and makes the result more useful later. You're building a record of what you tried and why.

Then you set up your variants. The control is your current page. Variant B is whatever you want to change. You can edit it with the visual editor or target elements with a CSS selector if you prefer.

The visual editor opens your actual website in the browser. Click on the element you want to change and type the new version. That's the whole interaction.

You can change headlines, button text, paragraph copy, and simple styling. The editor shows you exactly what your variant will look like before you publish it. If something doesn't look right, you can adjust it or start over.

For people comfortable with code, you can also target elements using CSS selectors and apply changes that way. Both approaches end up in the same place.

Once your experiment is live, SplitPea splits traffic evenly between control and your variant. Each visitor only ever sees one version, and they'll keep seeing the same one if they come back.

Your experiment page shows visitors, conversions, conversion rate, lift, and confidence. Confidence tells you whether the difference looks real, or whether you should wait before making the call.

You don't need to check it constantly. Some experiments take a few days, some take a couple of weeks. It depends on your traffic. SplitPea will tell you when there's enough data.

When confidence is high enough (SplitPea uses 90% as the threshold), you'll see a clear recommendation. You can end the experiment, apply the winning version to your site, and move on to the next thing.

If neither version wins, that's useful too. SplitPea will mark the result as inconclusive, and you'll know to try a different change or wait for more traffic.

Every experiment stays in your decision history. Over time, you get a useful record: what you tried, what happened, and what you decided to do next.