Experiments Answers you can act on.
Every rollout can double as an A/B test. Choose a goal metric and a guardrail, split the traffic, and Plinth shows the whole range of likely outcomes, not just a winner.
How it works
-
Pick a goal
Choose what success means, such as checkout completed or plan upgraded, from your events.
-
Split the traffic
Send users to each variant by percentage, sticky to each user across sessions and devices.
-
Read the result
See the likely effect and how sure you can be, in plain language, then ship the winner in one click.
The details
- Guardrail metrics
- Watch latency, errors and revenue alongside the goal, so a win never quietly costs you elsewhere.
- Bayesian or frequentist
- Choose the method your team trusts. Results say what they mean in a sentence.
- Holdouts
- Keep a small group on the old experience to measure the long-term effect of a launch.
const variant = plinth.variant('pricing-page', { user });
plinth.track('upgraded', { plan: 'team' });More of the platform
-
Feature flags
Put every change behind a flag and decide who sees it, without a deploy.
Learn more: Feature flags -
Scheduled rollouts
Go from 1% to everyone over days, with automatic pauses when errors rise.
Learn more: Scheduled rollouts -
Product analytics
Funnels, retention and cohorts, with every release marked on the chart.
Learn more: Product analytics
Try Experiments today
Free for small teams. Set up in minutes.