What Is a Good CTR on YouTube? Real Benchmarks

The short answer: 2% to 10%, but read the fine print

YouTube publishes an official benchmark, and it is wider than most people expect. According to YouTube’s own CTR FAQ, half of all channels and videos have an impressions click-through rate between 2% and 10%. New videos, channels less than a week old, and videos with fewer than 100 views can land well outside that range in either direction.

So if your CTR sits at 4% and you read somewhere that “good” starts at 8%, do not panic. The raw number means very little until you know where your impressions came from and how many of them you got.

Why your traffic source changes everything

The same thumbnail earns wildly different CTRs depending on where a viewer sees it:

Channel page: the highest CTR you will see. These viewers already chose you, so they click at high rates.
Search: high. The viewer typed a query and your video answered it. Well-matched search results routinely clear 8%.
Suggested videos: middle of the pack. You are competing next to a video the viewer just watched.
Browse (the homepage): the lowest. YouTube shows your video to broad, colder audiences, and most of them scroll past. A 3 to 4% CTR from Browse can be a strong result.

This is why comparing your overall CTR to another channel’s is mostly noise. A small channel whose views come from search will post a “better” CTR than a large channel getting millions of homepage impressions, even if the large channel’s thumbnails are stronger.

When a falling CTR is good news

Here is the counterintuitive part: on a video that takes off, CTR usually drops.

When a video performs well, YouTube pushes it to bigger and colder audiences on the homepage. Those viewers click at lower rates, so your percentage falls while your total clicks and views climb. A video at 4% CTR on 2 million impressions is doing far better than one at 9% CTR on 20,000 impressions.

Read CTR together with impressions, not on its own. Rising impressions with a gently falling CTR is the signature of a video being promoted. Flat impressions with a low CTR is the signal that your packaging needs work.

What to check if your CTR is under 2%

If a video sits below 2% after its first week or two, and impressions are not growing, the packaging is losing the impression. Work through these in order:

Is the thumbnail readable at phone size? Most impressions render smaller than a postage stamp. Squint-test it.
Is there one clear focal point? One face or object, not a collage.
Do the thumbnail and title repeat each other? They should each carry different information. If the text on the thumbnail restates the title, you wasted half your pitch.
Does the packaging match the search intent? A great thumbnail on a video nobody was looking for still gets skipped.

We covered the design side in detail in how to make clickable YouTube thumbnails.

CTR only pays off if people stay

YouTube warns about this directly: clickbait that wins the click but loses the viewer backfires. Videos with high CTR and low average view duration are less likely to get recommended, because the click told YouTube one thing and the instant drop-off told it another.

The practical rule: your thumbnail should make a promise the first 30 seconds of the video keeps. If you tease a result, show progress toward it early. A 6% CTR with solid retention beats a 12% CTR where half the audience leaves in the first minute.

Stop guessing: test your thumbnails

You do not have to debate which thumbnail is better. YouTube’s built-in Test & compare feature runs a true A/B/C test: upload up to three thumbnails on a long-form video, YouTube shows all of them to viewers at the same time, and it declares a winner based on watch time share. Not just clicks, watch time, which filters out clickbait wins automatically.

The bottleneck is having three genuinely different thumbnails to test. Making one decent thumbnail already takes most creators 30+ minutes, so almost nobody tests variations. That is where Thumbsmith earns its keep: describe the video once, generate several distinct concepts in minutes, pick the three strongest, and let the test settle the argument. Do that for ten uploads in a row and you will know more about what your audience clicks than any benchmark article can tell you.

Your CTR is not a grade. It is a feedback loop, and thumbnails are the input you control.

Make three thumbnails worth testing →