# A/B testing

Comparing randomly assigned versions of an experience to estimate the effect of a change.

Comparing randomly assigned versions of an experience to estimate the effect of a change.

## An example

A publisher randomly assigns eligible users to one of two offer-card layouts.

## Why it matters

Choose the outcome and guardrails before reading results, and avoid treating a small random difference as a reliable improvement.


## Related terms
- [Conversion rate](https://jolts.xyz/resources/glossary/conversion-rate)
- [Holdout group](https://jolts.xyz/resources/glossary/holdout-group)
- [Incrementality](https://jolts.xyz/resources/glossary/incrementality)

## Next step

[Explore advertising](https://jolts.xyz/advertisers) or [monetize your AI app](https://jolts.xyz/publishers).