# How to A/B test YouTube thumbnails

By HeyMark · Updated November 5, 2026

If you have several thumbnails for a video and want to compare them before choosing, you can run an A/B test in YouTube Studio. First check whether the video is eligible; then add up to three images and review the result based on watch time share.

## Check whether the video is eligible

You must set up the test on a computer, and your channel needs advanced features enabled. Shorts, scheduled live streams, and active Premieres are not eligible. A Premiere may qualify after it ends and becomes a long-form video. Private, made-for-kids, and age-restricted videos are also excluded.

YouTube Help lists the [A/B testing requirements and results](https://support.google.com/youtube/answer/16391400?hl=en).

## Set up a thumbnail-only test

1. Open [YouTube Studio](https://studio.youtube.com/) on a computer and go to **Content**.
2. Select the video you want to compare. In its details, find the **Thumbnail** box and click **A/B Testing**.
3. Choose **Thumbnail only**. This keeps the title out of the comparison; the feature also supports title tests, but this walkthrough compares images only.
4. Upload up to three thumbnails and click **Done**.

Do not change the thumbnail or title while the test is running. Either change stops it, so you will need to start a new test.

## Review the test result

To check progress, open the video in Studio, go to **Analytics**, select the **Reach** tab, and click **Manage test** in the A/B testing section. You can also find the result in the video's details.

YouTube compares watch time share, not click-through rate (CTR) alone. A test may identify a winner, show similar performance, or be inconclusive. If there is no clear winner, the first thumbnail remains the default, and you can choose manually afterward. “Inconclusive” means the data did not clearly distinguish an option; it does not mean the test failed.

Timing varies with impressions and how different the options are. A test can take a few days and may run for up to two weeks. If it ends without a winner, compare the images and choose the one that best represents the video instead of treating a small difference as certain.

To save the report with the period and filters you are comparing, see [export YouTube Analytics reports from Studio](https://heymark.ai/en/solutions/export-youtube-analytics-reports-studio/).

## Frequently asked questions

### Can I test a Short thumbnail or an active Premiere?

No. Shorts and active Premieres are not eligible. A Premiere may be tested after it ends and converts to a long-form video.

### What happens if I change a thumbnail during the test?

YouTube stops the test if you change the title or thumbnail. You will need to set up a new test to compare the options.

### Why is there no winning thumbnail?

YouTube may report similar performance or an inconclusive result when the differences between options or the available impressions do not clearly distinguish them. The first option remains the default, and you can choose manually.

## See it on your own accounts.

Connect Instagram, TikTok, Facebook, LinkedIn or YouTube and use HeyMark free for 7 days. Mark answers in your voice, and you approve what goes out.

[Start free](https://app.heymark.ai) · [See pricing](https://heymark.ai/en/pricing/)

## Developer and agent resources

- [HeyMark MCP documentation](https://heymark.ai/en/mcp/)
- [llms.txt](https://heymark.ai/llms.txt)
- [Full site content for language models](https://heymark.ai/llms-full.txt)
- MCP protocol endpoint: `POST https://mcp.heymark.ai`
- OAuth protected-resource metadata: [/.well-known/oauth-protected-resource](https://mcp.heymark.ai/.well-known/oauth-protected-resource)
- MCP server card: [/server-card](https://mcp.heymark.ai/server-card)