Thumbnail click-through rate
Which thumbnails convert best after adjusting for reach?
Click-through rate falls as impressions grow, because the marginal impressions go to less-targeted viewers. A viral video at 3% CTR on a million impressions is not worse packaging than a niche video at 10% on a thousand.
Odinstat fits the expected CTR for each level of impressions on your channel and ranks thumbnails by how far above or below that expectation they landed.
Why raw CTR misleads
YouTube Studio reports impressions click-through rate per video. Sorting by it will almost always put your smallest videos on top, because a thumbnail shown mostly to subscribers and search visitors converts far better than one pushed to millions of browse-feed strangers. The number is real; the ranking is not about the thumbnail.
Reach-adjusted CTR
The relationship between CTR and impressions is approximately log-linear. Odinstat fits that line across your own videos, giving an expected CTR for any impression count on your channel. Each video’s residual, actual minus expected, is the reach-adjusted score. A positive residual means the thumbnail and title earned more clicks than a video shown that widely usually does for you.
The same reach-adjusted CTR is the click axis of Odinstat’s packaging-versus-content verdict, which separates “people did not click” from “people clicked and left”.
What YouTube lets us see
Impression data comes from the YouTube Reporting API and only exists from roughly 30 to 60 days before the day a channel connects to Odinstat. Older videos therefore have no impression history and cannot be scored; that is a YouTube limit, not a gap Odinstat fills with an estimate.
A CTR computed from a handful of impressions is noise, so a video needs at least 100 tracked impressions before Odinstat will report or rank its CTR.
What you need
At least 6 long-form videos that each have at least 100 tracked impressions. Shorts are excluded (the Shorts feed does not use thumbnails the same way). Separately, impression history begins about 30–60 days before your channel connected.
What this is not
- A/B testing thumbnails before you publish. Odinstat measures what happened to thumbnails you already shipped.
- Predicting CTR for a channel you do not own.
Odinstat runs this on your own channel, read-only, with a 14-day free trial from $10/month.