YouTube thumbnail tester:
compare variants before you publish.
Upload two or three versions of a thumbnail and one title. See them side by side at feed size, blurred, in grayscale, tiny and under the duration badge, with simple measured checks for each. It shows how each version reads. It does not predict clicks.
Add your variants
Two or three versions of the same thumbnail, and the title you will publish with. Images that are not 16:9 are cropped to 16:9 for the views.
Add at least one image to open the comparison views. Two or three is where it gets useful: the views show them side by side, and the measurements compare them.
Five ways to look at the same thumbnail.
Each view answers one question. None of them tells you which thumbnail will win. They show you where a variant is weak before you spend a real test on it.
- Feed test
- Your variants sit among neutral placeholder tiles at desktop and phone width. Look away, look back, and notice which one your eye reaches first. Shuffle the order, because position changes what gets noticed.
- Squint test
- Raise the blur until the picture is a smear. What still stands out, whether a shape, a face or a block of colour, is what a fast scroller sees. If nothing stands out, the composition is not doing much work.
- Grayscale test
- Colour can hide weak contrast. In grayscale, check that the subject separates from the background and that any words are still easy to read.
- Tiny test
- Sidebar and small-screen thumbnails are far smaller than in the feed. Check that the subject and the words are still recognisable at about 168, 120 and 84 pixels wide.
- Badge test
- The duration badge sits over the bottom-right corner of every thumbnail. Anything important there is hidden, and a longer time such as 1:02:45 covers more than 3:15.
How to A/B test thumbnails on YouTube properly.
The tester above checks how a thumbnail looks and reads. To find out which version people actually click, you need real viewers, and YouTube has a built-in way to get them.
What YouTube's own test does
- What it is called
- A/B test titles & thumbnails, in YouTube Studio. It was first released as Test & Compare, and many creators still call it that.
- Where you run it
- In YouTube Studio on a computer. Advanced features must be turned on for your channel.
- What you can compare
- Up to three titles, thumbnails, or title-and-thumbnail combinations on one video.
- How it decides
- By watch time share, not by click-through rate. YouTube says it optimises tests for overall watch time over other metrics, such as click-through rate.
- How long it takes
- A few days, and up to two weeks.
- What you get back
- One of three results: Winner, Performed same or Inconclusive.
- What it does not cover
- Shorts, scheduled live streams and Premieres, among others. YouTube Help has the full list.
- Image quality
- If any thumbnail in the test is below 1280 × 720, YouTube downscales every option to 480p.
Checked against YouTube Help on 29 September 2026. YouTube changes this feature, so read the current Help page before you rely on a detail.
How to run a test that teaches you something
Change one thing between variants: the crop on the face, the words, the background colour. If all three are completely different, a win tells you which package worked but not why.
Drop a variant that is unreadable when small, blends into the feed, or hides its point under the duration badge. Do that before you spend a test on it.
Below that, YouTube downscales the whole test to 480p. Thumbnail size guide.
Stopping early because one option looks ahead is the classic way to fool yourself. Let it finish.
It is based on watch time share. Performed same or Inconclusive means YouTube saw no clear difference, so choosing on taste is fair. Judge click-through rate separately, on real impressions, before and after the change.
The lasting value of a test is the rule you carry to the next video, not only the winner of this one.
Now let real viewers decide.
This page cannot tell you which thumbnail gets clicked. That depends on your audience, your title and what else is in their feed. Two things can: YouTube's own A/B test, and your real click-through rate (CTR), which is the share of people who saw a thumbnail and clicked it.
Connect your channel and Base.Tube's free audit shows real impressions and CTR for your videos, including how CTR changed before and after you swapped a thumbnail. Need more versions to test? Generate them in the Studio.