> ## Documentation Index
> Fetch the complete documentation index at: https://docs.mention-me.com/llms.txt
> Use this file to discover all available pages before exploring further.

# A/B testing troubleshooting

> Common issues when setting up, running and concluding an A/B test, and how to resolve them.

## Setup Issues

| Issue                                                                                               | What to do                                                                                                                              |
| --------------------------------------------------------------------------------------------------- | --------------------------------------------------------------------------------------------------------------------------------------- |
| "The A/B test variants are identical. Differentiate them to run a meaningful A/B test."             | Edit Variant B so it differs from the control. See [manage an A/B test](/knowledge-v2/referral/campaigns/ab-testing/manage-an-ab-test). |
| "This A/B test is missing required settings. Set the test metric and what happens on each outcome." | Reopen the wizard via **Adjust test settings** and complete the metric and outcome steps.                                               |
| The promotion option is missing                                                                     | A time-limited promotion cannot run an A/B test. Use a standard version.                                                                |

## I Can't Edit the Test

You cannot edit a version while its A/B test is live, so set the metric, speed and outcomes before you launch. The **control** variant is always view-only; edit the other variant.

## The Test Is Taking a Long Time

* A test runs to a maximum of 8 weeks (occasionally up to 12 where a return or cooling-off window applies). Most winners are found within about 4 weeks.
* For low-volume audiences, choose **Share rate** as the test metric, or pick a faster speed (which uses a lower confidence target). With very low volumes, it is often better to follow referral best practices than to split limited traffic between variants.

## There's No Clear Winner

If the test reaches its maximum duration without a clear winner, Mention Me applies the outcome you chose in the **When it concludes** step (for example keep the control, or allocate to the best performer). You can review the results and [switch traffic](/knowledge-v2/referral/campaigns/ab-testing/monitor-an-ab-test) to the other variant once.

## Still Stuck?

[Contact support](https://help.mention-me.com/hc/en-gb/requests/new).