Knowledge Bank
3 min read
The SEO checker, and why tags are worth ten points
Five checks on any public video, weighted the way they actually matter rather than the way people fear.
Not sure if Package is your weak phase? The free 3-minute workflow audit will tell you.
Most of what decides whether a video works happened long before the description box: the idea, the hook, whether anybody wanted it. The SEO checker looks at what is left, which is the part you can still fix in five minutes without reshooting anything. Paste any public video, yours or somebody else's, and it scores five things out of a hundred.
The weighting is the whole point
| Check | Points | What it reads |
|---|---|---|
| Title | 25 | Length and casing. Search cuts it around sixty characters |
| Description | 25 | Total length, and what sits in the first 150 characters |
| Thumbnail | 20 | Whether a 1280x720 image exists at all |
| Chapters | 20 | Timestamps present, and formatted the way YouTube requires |
| Tags | 10 | Usage against the 500-character field |
Tags carry ten points because that is roughly what they are worth in practice. A checker that scored them like a ranking factor would send you off to spend an afternoon on the least valuable field on the page, feeling productive the whole time.
What each check is really asking
The title check is about the tail. Search truncates around sixty characters, and the tail is where most people put the reason to click, so a long title often loses exactly the part doing the persuading.
The thumbnail check is blunter than it looks. A missing 1280x720 image almost always means YouTube auto-picked a frame, and an auto-picked frame is a random moment of your face mid-word.
The description check reads the first 150 characters separately from the rest, because that slice is what shows before Show more. Everything below it is for people who already decided to expand.
Read a low score correctly
This distinction matters more on somebody else's video than your own. Run a video that clearly outperformed and you will sometimes get a mediocre score, which tells you the packaging was not why it won. Run a video that flopped and score well, and you have ruled out the cheap explanations.
Checks that do not apply drop out
Chapters on a two-minute video would be a rule nobody could satisfy, so that check leaves the total entirely rather than costing points. The score rebases on what was actually achievable.
That sounds like a detail, and it changes how you should read the number. A 78 on a video with four applicable checks is a different result to a 78 on five, so look at which lines are missing before you compare two videos.
Where it fits in the run
Use it as the last pass on a private draft, after the title is locked and the description is written. Everything it flags is editable in Studio in a couple of minutes, and doing it before the video goes public means you fix things without touching a live listing.
Running it on old uploads is the other use. Pick the five videos still pulling views and check them. Old videos keep earning, and the packaging on them was often written when you knew less.
Keep reading
For the title itself, picking the title covers what the length check cannot. The first two lines are the whole argument in the description structure guide, and the tag score makes more sense alongside what tags are still worth in 2026.