Create and manage experiments

Updated Nov 8, 2026 Based on Salesforce docs updated 2026-08-27

An experiment compares personalization variants against a control by splitting visitor traffic between them, so you can tell which one actually performs better instead of guessing. Run one before making a change permanent.

Set up an experiment

  1. From Personalization, create an experiment on the decision or recommender you want to test.
  2. Keep the current experience as the control, then add one or more variants (for example, a different recommender or point configuration).
  3. Set the traffic allocation that splits visitors across the control and each variant.
  4. Choose the success metric the experiment should optimise for (for example, clicks, conversions, or revenue).

Run and act on results

  1. Start the experiment so it begins serving the control and variants to live traffic.
  2. Let it run until it reaches enough visitors and time to be meaningful, not just until a variant looks ahead.
  3. Review the results per variant against your chosen success metric.
  4. Promote the winning variant to 100% of traffic, or end the experiment if no variant clearly wins.
Experiment configuration with control, variants, traffic split, and success metric
Screenshot: Create an experiment

Discussion

Comments are not configured for local development. They appear once the site owner connects GitHub Discussions (Giscus).

Support SF Guide

SF Guide is a free, independent guide to Salesforce Personalization, built and maintained in the open. If it saved you time on an implementation, a small contribution helps keep it current and ad-light.