![A/B Test YOUR YouTube Thumbnails For FREE! [A/B Testing Tutorial] thumbnail](https://img.youtube.com/vi/U0S-TuEaLwE/mqdefault.jpg)
YouTube's FREE Thumbnail A/B Testing Tool!
38sReveals a new free tool from YouTube that creators have been waiting for, promising to solve the guesswork in thumbnail selection.
▶ Play ClipAI Summary
YouTube has released a free thumbnail A/B testing tool that allows creators to test up to three thumbnail options to see which performs best based on watch time. This tool helps creators move beyond guessing and use data to optimize their thumbnails for higher click-through rates and viewer retention.
Chapters
[00:18]
Importance of Thumbnails
90% of best-performing YouTube videos use custom thumbnails.
[00:33]
Testing Over Guessing
The best thumbnail is the one your audience likes, not necessarily the one you like. Mr. Beast creates up to 50 thumbnails per video and tests them.
[01:04]
How A/B Testing Works
Upload thumbnails, YouTube shows each to different viewers, tracks performance, and automatically updates to the winner based on watch time.
[01:28]
Watch Time as Metric
YouTube picks the winner based on watch time, not just clicks, ensuring viewers actually watch the content.
[02:05]
Setup Requirements
Available on desktop, requires advance verification (free). Not available for made-for-kids or age-restricted content.
[02:21]
Setting Up a Test
For new uploads: click create, upload video, find A/B testing option near title area. For existing videos: go to edit page, click A/B testing in title or thumbnail section.
[02:47]
Test Options
Three options: title only, thumbnail only, or title and thumbnail. Minimum two thumbnails needed; can add a third for A/B/C testing.
[03:20]
Making Different Thumbnails
95% of click decisions are subconscious. Make thumbnails noticeably different in colors, expressions, angles to avoid prolonged tests.
[03:55]
Using Adobe Express
Adobe Express allows quick creation of multiple thumbnail versions with brand colors and fonts, accessible across devices.
[04:56]
Interpreting Results
Results can take up to 2 weeks. If no clear winner, check analytics under reach tab or use Ask Studio feature. If insufficient data, switch to new thumbnail and study results in a week.
YouTube's free A/B testing tool empowers creators to make data-driven thumbnail decisions, improving click-through rates and viewer retention. Consistent testing helps creators learn what resonates with their audience.
Clickbait Check
95% Legit"Title accurately describes the free A/B testing tool and tutorial; content delivers exactly what's promised."
Mentioned in this Video
Tutorial Checklist
4
04:39
Click 'Add thumbnail' to upload your first thumbnail option, then again for the second (and third if desired).
6
04:56
Wait up to 2 weeks for results. YouTube will automatically apply the winning thumbnail based on watch time.
7
05:01
If no clear winner, check analytics under 'Reach' tab or use 'Ask Studio' feature. Manually choose the better-performing thumbnail.
Study Flashcards (8)
What percentage of best-performing YouTube videos use custom thumbnails?
easy
Click to reveal answer
What percentage of best-performing YouTube videos use custom thumbnails?
easy
Click to reveal answer
90%
00:18
What metric does YouTube use to determine the winning thumbnail in A/B testing?
medium
Click to reveal answer
What metric does YouTube use to determine the winning thumbnail in A/B testing?
medium
Click to reveal answer
Watch time, not just clicks.
01:28
What are the three options available when setting up an A/B test in YouTube Studio?
medium
Click to reveal answer
What are the three options available when setting up an A/B test in YouTube Studio?
medium
Click to reveal answer
Title only, thumbnail only, and title and thumbnail.
02:47
How many thumbnail options are required for an A/B test?
easy
Click to reveal answer
How many thumbnail options are required for an A/B test?
easy
Click to reveal answer
At least two.
03:01
What is the maximum number of thumbnails you can test at once?
easy
Click to reveal answer
What is the maximum number of thumbnails you can test at once?
easy
Click to reveal answer
Three (A/B/C testing).
03:06
Why should thumbnails be noticeably different in A/B testing?
medium
Click to reveal answer
Why should thumbnails be noticeably different in A/B testing?
medium
Click to reveal answer
Testing thumbnails that are too similar can cause the test to run longer because there's not enough difference to pick a winner.
03:35
What percentage of click decisions happen before people consciously think about it?
hard
Click to reveal answer
What percentage of click decisions happen before people consciously think about it?
hard
Click to reveal answer
95%
03:26
What should you do if YouTube says there's not enough data to declare a winner?
hard
Click to reveal answer
What should you do if YouTube says there's not enough data to declare a winner?
hard
Click to reveal answer
Go to that video's analytics and look under the reach tab for more data, or use YouTube's Ask Studio feature.
05:06
💡 Key Takeaways
📊
90% Use Custom Thumbnails
Statistic from YouTube emphasizing the importance of custom thumbnails for performance.
00:18
⚖️
Test, Don't Guess
Key principle that the best thumbnail is determined by audience preference, not personal taste.
00:33
🔧
Watch Time as Success Metric
YouTube uses watch time to select winning thumbnail, ensuring it attracts viewers who actually watch the content.
01:28
📊
95% Subconscious Decisions
Highlights the subconscious nature of click decisions, reinforcing the need for testing.
03:26
💡
Continuous Testing
Emphasizes that consistent testing leads to better understanding of audience preferences.
05:35Full Transcript
[00:00] YouTube thumbnails matter a lot when it
[00:03] videos. Adobe Express sponsored this video so I
[00:08] which is their own free thumbnail testing tool, so
[00:13] YouTube built this tool because thumbnails are one
[00:18] clicks through to watch your video. YouTube even
[00:24] use custom thumbnails. The problem myself and
[00:28] hard to reliably predict what thumbnail will
[00:33] thumbnail isn't always the one you like, it's the
[00:38] it is to test it. What looks great for me might
[00:44] has a team that creates up to 50 thumbnail options
[00:50] tracks when videos change their thumbnail. Mr.
[00:55] thumbnails, but you might need more than one. I
[01:00] be testing," and I'm glad that YouTube has
[01:04] A/B testing YouTube thumbnails works. You add the
[01:10] each option to different viewers and tracks which
[01:15] though I've seen them sooner, and when the
[01:19] update your video to the winning thumbnail. Of
[01:24] winning thumbnail has been selected. YouTube also
[01:28] watch time. Great YouTube thumbnails don't just
[01:33] what the video is about so they don't waste time
[01:38] picks the winner based on watch time, not just
[01:43] one that people click on, it's the one that brings
[01:48] I've been recommending YouTube thumbnail A/B
[01:51] at TubeBuddy, it was one of the biggest claims
[01:56] starting or you've been A/B testing for years
[02:00] stop wondering what works and start knowing.
[02:05] is currently available on desktop and does require
[02:10] don't worry, it's free and only takes a few
[02:15] is not available for made-for-kids content or age
[02:21] To set up an A/B test, you can test it on either
[02:26] click create in the top right-hand corner, then
[02:31] see the A/B testing option near the title area and
[02:37] go to the video's edit page. For titles, click
[02:42] scroll down to the thumbnail section and click
[02:47] When you click here, you'll see 3 options,
[02:52] thumbnail. These are all worth exploring, and I
[02:57] can subscribe to be notified when those drop.
[03:01] only. You only need 2 thumbnail options for
[03:06] third thumbnail. That's sometimes called "A/B/C
[03:11] just with 3 options instead of 2. Now, I
[03:15] it's hard enough to make one good thumbnail, how
[03:20] you and I, we can't reliably predict what will
[03:26] happen before people even consciously
[03:30] when you're making your second or third thumbnail,
[03:35] different expressions, different angles. YouTube
[03:40] can cause the test to run longer because there's
[03:44] people in our KDCC Discord server say they've
[03:50] punctuation on the text of the thumbnail. Sadly,
[03:55] make multiple thumbnails without spending hours on
[04:00] makes it easy. That's why I like Adobe Express.
[04:04] and the fonts once. Then, from there, you can
[04:10] and create multiple versions without starting from
[04:15] phone and finish it on your computer, or the other
[04:20] it on your phone. I was using Adobe Express before
[04:25] and use it. I even made versions of the thumbnails
[04:31] try it out, head to adobe.com/express or use my
[04:36] YouTube channel. Once you got your thumbnails
[04:40] thumbnail to upload your first option, then click
[04:46] to do an A/B/C test, add a third one. Once they're
[04:51] actually save your changes on the video. I've seen
[04:56] can take up to 2 weeks, and sometimes YouTube
[05:01] YouTube picks a winner, go with that. If YouTube
[05:06] or the YouTube thumbnails perform the same, go to
[05:11] tab for more data. You can also use YouTube's own
[05:16] data. Do your best to pick the thumbnail that's
[05:21] based on the data that you do have. If there
[05:26] testing, consider switching to the new thumbnail
[05:30] there was a reason you wanted to test in the first
[05:35] The more you test, the more you learn what works
[05:39] more A/B testing tips, an Adobe Express affiliate
[05:44] KDCC Discord server if you need more help. If
[05:48] out that video on screen. I'm Andrew Kan, and if I