YouTube Analytics: The 5 Metrics That Actually Matter

Open YouTube Studio and you get forty numbers before your coffee is cold. Watch time, unique viewers, average percentage viewed, stickiness, shares, likes-per-view. Most of them will not change a single thing you do this week. Five of them will. Here is what to read, in order, and what each one tells you to fix.

Vanity metrics vs decision metrics

A vanity metric feels good and tells you nothing to do next. Total views over the channel’s lifetime, subscriber count on its own, likes. They go up when you do well, but they do not point at what to change. A decision metric does one job: it names the exact thing that is capping the video, so you know what to fix on the next one. The five below are decision metrics.

1. Impressions: are people even seeing it?

Impressions count how many times YouTube put your thumbnail in front of someone, in search, suggested, or home. Low impressions on a video that should have demand usually means one of two things: the topic itself has thin search volume, or YouTube has not found an audience to show it to yet (common in the first 48 hours). If impressions never grow past a few hundred, the problem is upstream of the video. It is the topic.

2. Click-through rate: does the packaging earn the click?

CTR is clicks divided by impressions. It answers a narrower question than impressions: of the people who saw the thumbnail and title, how many cared enough to click? A healthy CTR for most niches sits somewhere in the 4 to 10 percent range, though it varies a lot by niche and by how loyal your existing audience is. If impressions are fine but CTR is low, the video is getting shown and getting ignored. That is a packaging problem, not a content problem. The fix is the title and thumbnail, not the edit.

3. Average view duration: does the video keep the promise?

Average view duration (AVD) is how long people actually stay, in minutes and seconds. Pair it with average percentage viewed, which is AVD as a share of total runtime. If CTR is strong but AVD collapses in the first 30 seconds, the title opened a gap the video did not close fast enough. People clicked because you promised something. If the video takes a minute to get there, they leave before you pay it off. This is also where a slow intro, a logo animation, or a “before we start” preamble quietly costs you the video.

4. Returning viewers: are you building an audience or one-off hits?

The returning viewers percentage in the audience tab tells you whether people who watched are coming back for the next video, or whether every upload is starting from zero with strangers. A channel with strong single-video numbers but a low returning-viewer share is not actually building anything yet. It is making individual hits. The fix is usually a narrower, more consistent promise: the same kind of video, for the same person, often enough that they learn what to expect from you.

5. Subscribers gained per 1,000 views: does the video convert into an audience?

This one gets skipped, but it is the clearest signal that a video is doing its full job, not just getting watched. A video with a great AVD but almost no subscriber conversion is entertaining people who have no reason to come back. Compare it across your last 10 videos and you will usually find one or two formats that convert noticeably better than the rest. Make more of those.

Read them together, not one at a time

The five metrics are a diagnostic chain, and the fix depends on where the chain breaks:

1. Low impressions, whatever else is happening → the topic itself lacks demand or is too new to have found its audience yet. Check the topic before you touch the video.
2. High impressions, low CTR → the title or thumbnail is not earning the click. Rewrite the packaging, not the video.
3. High CTR, AVD drops in the first 30 seconds → the hook does not deliver the promise fast enough. Cut the preamble.
4. Good AVD, low returning viewers → the format is not repeatable enough for people to expect a next one. Narrow the promise.
5. Good AVD, low subscriber conversion → the video entertains but gives no reason to follow. Add a clear reason to stick around, on screen, not just in the description.

Chase the wrong number and you burn hours fixing something that was never broken. A video with plenty of impressions and a weak CTR does not need a better hook. It needs a better title.

A faster read on all five

Pulling these five numbers for your last 10 to 15 videos by hand works, it just takes longer than it should. TubeHunter’s free channel breakdowns line up the packaging and retention side by side so you can see which lever is actually capping a video before you spend a weekend re-editing it. If the packaging is the question, the title scorer is the faster first check.

The one-line version: impressions tell you if anyone saw it, CTR tells you if the packaging earned the click, AVD tells you if the video paid it off, and returning viewers plus subscriber conversion tell you if you are building an audience or just racking up one-off views. Fix them in that order, not whichever one is lowest on the page.