Deploy a winning variation

  • Updated
  • Optimizely Feature Experimentation

If you see a winning variation on your Experiment Results page, you may decide to deploy that variation to route incoming traffic to the winner to increase conversions. Instead of relying on developer resources to implement the winner in your source code, you can use Optimizely Feature Experimentation to deploy the winning variation instantly. See the walkthrough to see how to conclude an experiment in action.

Optimizely counts any user ID it sees as a MAU, see Monthly Active Users (MAUs) for information.

The Concluded status changes the experiment status and displays it as Concluded in the experiment list. Additionally, concluding a rule removes it from the datafile. Setting a rule to Concluded stops serving the variations unless you select Conclude and deploy. See Flag and rule lifecycle management for information.

Conclude and deploy a variation

When you conclude a rule, you can select the outcome of your rule and add a summary of what you learned from your hypothesis. This helps track whether the experiment had a positive, negative, or inconclusive result.

Setting a rule to Concluded ends the experiment and removes it from the datafile. Feature Experimentation stops serving variations unless you choose Conclude and deploy.

After concluding your rule, you cannot modify that rule. To conclude a rule and deploy a variation,

  1. Go to Flags > select your flag.
  2. Select the environment.
  3. Click More options (...) on the rule.

  4. Select Conclude.

  5. Select an option for the Results Outcome.
  6. (Optional) Add a brief conclusion about what you learned about the experiment.

  7. Toggle Conclude and deploy. If the flag is temporary, you can continue to expose users to the winning variation, which can help avoid unnecessary complexity in your codebase. However, if the flag is long-lived, this approach may not be appropriate.

  8. Click Next.
  9. Select the variation that performed the best from the Variation drop-down list to direct traffic to it.

  10. (Optional) Adjust the Audience and Traffic allocation percentage.

  11. Click Deploy.

After concluding the experiment and deploying a rule, a Targeted Delivery rule is created for the selected variation.

The Optimizely Experiment Results page displays a message that the experiment was concluded and deployed.