YouTube Launches New Thumbnail Testing Tool to Boost Your Views

Are you struggling to make thumbnails that stop viewers in their tracks and inspire them to click on your videos? If so, you’re going to love this update geared toward creators of all sizes.

YouTube just announced a new tool, Test and Compare, that lets you upload multiple thumbnails and test them with different audiences. The results are pretty straightforward: Whichever graphic has the highest share of watch time wins the attention war. Best of all, the tool is 100% free, and all YouTube creators will have it by early 2024.

This is a game changer for small channels, as it’s never been harder to stand out on YouTube and get noticed by your target audience. Thumbnails do a lot of that work, from grabbing viewers' attention to providing an emotional snapshot of the video beforehand. So, getting to test multiple thumbnails is a fantastic way to increase your views.

Accessing the Test and Compare Tool

YouTube is gradually rolling out the Test and Compare tool, but you might have early access. To find out, upload a video as you normally would and pay close attention when it’s time to upload a thumbnail. If you have the feature, you’ll see an option labeled “Test and Compare” below Reselect and Upload File.

Does ‘Test and Compare’ Work for all Content Types?

For now, you can test and compare thumbnails on long-form videos, completed livestreams, and podcasts. Certain types of content, like those aimed at kids or mature audiences, are excluded from using this tool.

YouTube Thumbnail Testing: How to Measure Performance

YouTube’s new tool follows the rules of A/B testing. It allows you to upload multiple thumbnails (two or three) for the same video. YouTube then displays these thumbnails to different segments of a creator’s audience, each one getting an equal share of exposure.

The goal is to determine which thumbnail is more effective in terms of viewer engagement and performance. And YouTube is using a new metric, “watch time share”, as its benchmark of choice. That’s not to be confused with average view duration, which is the average amount of time viewers spend watching a video.

Click-Through Rate vs. Watch Time Share

From a testing perspective, the share of watch time each thumbnail earns is what YouTube describes as watch time share. For example, if you test and compare two thumbnails, thumbnail A might accumulate 60% of the watch time, while thumbnail B collects 40%. In this case, thumbnail A is the winner and most likely to boost your views.

But why is YouTube using watch time to test thumbnail performance instead of clicks? After all, the click-through rate measures how attractive a video is on search pages and other display areas.

It turns out that click appeal isn’t enough. In the YouTube Studio, the platform explains why watch time is a better predictor of satisfaction than clicks, and how thumbnails reinforce that.

“Great thumbnails don’t just get viewers to click,” YouTube states. “They also help viewers understand what the video is about, so that they can make informed decisions about what to watch.”

Insights from Thumbnail Testing

It’s awesome that YouTube is helping creators choose the best thumbnails for their videos. But the real question is, does it work? Do you get more views after testing thumbnails and using the best one?

Yes!

In a simple experiment, we tested three thumbnails for this video about new AI developments on YouTube.

The AI robot image was winning by a small margin, so we dropped the worst-performing thumbnail and let the stronger options compete against each other. Eventually, we got better results: The AI robot was leading the watch time share with a 59.7% split.

So, we dropped the other thumbnail and used the strongest performer (AI robot image). In the end, the video got three times the amount of views that we normally get.

New to Thumbnail Testing? A Word of Advice

YouTube’s new thumbnail tester is a major step up when you want to optimize your video’s appeal. At the same time, not everyone has the time or resources to create two or three thumbnails per video. If that’s your situation, do small tests instead of large, time-consuming experiments. Make tiny design shifts, like testing different fonts, colors, or simple backgrounds. Even those small tweaks can make a difference in your viewer’s eyes.

This video has more details and advice, so give it a quick watch to cement your knowledge.

But even without a thumbnail test, are you designing images your viewers can’t resist?

Here are 12 thumbnail styles that people love on YouTube!