Platform.

Everything the product does, on every paid tier.

What is actually in the product, rather than what the pricing page implies. Every one of these is on every paid tier; the tier decides how many seats, environments and evaluations you get, not which features are switched on. If something here does not do what you need, say so and we will tell you straight whether it is on the roadmap.

Talk to sales

Targeting rules

Generally available

Decide who sees a change by the attributes you already have: account id, plan, region, app version, or anything else you pass in. Rules are read top to bottom and the first match wins, so what a user gets is something you can work out by reading, not by guessing. Changing a rule takes effect on the next evaluation, which is milliseconds, not a deploy.

Percentage rollouts

Generally available

Move a flag from one percent to a hundred in whatever steps you like, by hand or on a schedule you set once. The split is sticky per user, so somebody who saw the new thing this morning still sees it this afternoon, and it stays sticky when you raise the percentage rather than reshuffling everyone.

Experiments

Generally available

Point two variants at the same flag, send traffic to each, and read which one moved the metric you named up front. You get the exposure counts, the difference and how confident that difference is, written in a sentence rather than left as a table for you to interpret.

Audit and export

Generally available

Every flip, every rule change, who made it, when, and what the value was before. Searchable in the app, streamed to a webhook, or exported as newline delimited JSON for whatever you keep logs in. Retention runs as long as your plan allows, and Enterprise sets its own.

Also planned

  • Edge evaluation

    Planned
  • Infrastructure as code provider

    Planned
  • Scheduled flips

    Set a flag to go on at a time you pick, so nobody has to be awake for it.

    Planned