13 built-in conditions. Combine them with AND and OR, nested as deep as you need. Save the rule once, then point every experiment and every feature flag at it.
There is no rules engine to learn, and ordinary rules need no code at all.
Start freeEvery one of these is built in. Pick a condition from the list, fill in the values, and that is your rule. Nothing to write, nothing to install. The groups below just make the list easier to read, and you can mix conditions from any of them. Most targeting questions need one or two. The rest are here for when the question gets harder.
What the page already knows: where the visitor is, what they are browsing on, and how they arrived.
Who the visitor is: your own cookies, and whether they have been here before.
Anything your site passes in, plus a JavaScript condition for the cases a list cannot cover.
What is in the basket, and what this shopper has bought before. These are available on an audience saved to a single project.
Each condition is one test. Group tests with AND when they all have to be true, or with OR when any one of them will do.
Groups sit inside other groups, so a rule can be as specific as the question you are asking. There is no depth limit to work around.
The builder shows the whole tree at once. You can read who is in and who is out without holding the logic in your head.
The usual alternatives are worse: a rule you cannot quite express, or one line of code that only the person who wrote it can explain six months later.
None of this needs a developer. Whoever knows the campaign can build the rule that matches it, and read it back to a colleague without translating anything.
Give a rule a name and it becomes a saved audience. Any experiment can target it, and so can any feature flag rule.
On a feature flag the same audience does two jobs: it decides who a rollout reaches, and who takes part in an A/B test rule.
Everything points at one definition instead of its own copy, so there is one place to look and one place to change it.
Edit the audience and every experiment and flag using it picks up the new version. You never have to hunt down copies that were meant to match and quietly stopped.
You write the rule once. Where it is checked depends on where the experiment or the flag lives, and you get the same answer either way.
For experiments on your site, the rules travel with the page and are checked there. The visitor waits for nothing.
For anything decided on your own servers, the same rules run inside your code. Pass what you already know about the user and get the same answer.
Free for your first 10,000 visitors a month. No card, and not a trial.