This article helps you:
Add recommendation and guardrail metrics to your experiment
Create new metrics from scratch, and edit existing metrics
An experiment can’t tell you anything without events to track. Adding metrics to your experiment occurs in the Goals segment of the experiment design panel. Here, you’ll tell Amplitude Experiment what you want your recommendation metric to be, as well as define any secondary metrics. A recommendation metric is any metric that you use to determine whether your hypothesis is accepted or rejected, and therefore, whether your experiment has succeeded or failed.
Your recommendation metric is important to understand the success of your experiment, so it’s important to choose the right one. If you’re not experienced in A/B testing, it can be hard to know which metric to choose. To help you create a successful recommendation metric, keep in mind the following:
One common mistake is defaulting to a revenue metric. This happens when your variant introduces a change that's separate from the metric you’ve selected. If your variant changes how your product page looks and functions, choose a metric on that page as your recommendation metric instead of a revenue metric that might not be visible for several more steps down the funnel.
Experiment lets you define multiple metrics when running an experiment. Unlike a recommended metric, non-recommended metrics aren’t required, but they're often helpful. They not only improve the quality of your analysis, but help evaluate whether it’s even worthwhile to roll out your experiment at all.
The duration estimator estimates the time and sample size you need to achieve significant results in your experiment, given your metric settings. Amplitude Experiment pre-populates reasonable industry defaults based on historical data, but you can adjust the confidence level, statistical power, minimum detectable effect, standard deviation, and test type as needed.
You can create a new metric if none of the standard metrics meet your needs.
In your experiment, open the Design Experiment panel, or the Analysis Settings, and choose the exposure event. When a user trigger this event, Amplitude Experiment buckets them into the experiment. The Amplitude exposure event is the most accurate and reliable way to track user exposures to your experiment’s variants, so you should use that if possible.
Amplitude sends the Amplitude exposure
event when your app calls .variant()
. It sets the user properties Amplitude Experiment uses to conduct its analyses. When you use the Amplitude exposure event, you can be certain your app triggers the event at the correct time.
You can select a custom exposure event instead. Click Custom Exposure, then Select event. There's a much greater risk of triggering a custom exposure event at the wrong time, which can lead to a sample ratio mismatch.
For more information, go to this article about exposure events.
After you know your goal, you can define your experiment's audience.
April 30th, 2024
Need help? Contact Support
Visit Amplitude.com
Have a look at the Amplitude Blog
Learn more at Amplitude Academy
© 2025 Amplitude, Inc. All rights reserved. Amplitude is a registered trademark of Amplitude, Inc.