YouTube Analytics Explained: Every Metric You Need to Track
Published July 2026
YouTube Analytics shows you more data than any creator actually needs to act on. The problem isn't a shortage of numbers — it's knowing which numbers point to real levers you can pull, and which ones are just noise that changes weekly without meaning anything. This guide covers every metric worth understanding and tells you directly which ones should drive your decisions.
Overview metrics: the channel-level snapshot
Views is the total number of times your videos were watched (minimum 30 seconds for most content). It's the most visible number and the least useful in isolation — views without context tell you nothing about whether your channel is healthy. A video with 10,000 views from the wrong audience can grow your channel less than one with 500 views from exactly the right audience.
Watch time (hours) is the total hours viewers have spent watching your content. This is the metric that counts toward your YouTube Partner Program eligibility threshold (4,000 hours in the past 12 months). Track it week over week to see your rate of accumulation. In YouTube Studio → Analytics → Overview, you can set the date range to the last 365 days to see your running YPP total.
Subscribers shows net subscriber change (gains minus losses). The gross number matters less than the trend: are you gaining more per video over time? Subscriber count by itself is vanity — a channel with 500 active, engaged subscribers is healthier than one with 5,000 disengaged ones.
Per-video performance metrics
Impressions is the number of times your thumbnail was shown to viewers on YouTube (in their feed, search results, or recommended section). Low impressions means YouTube isn't distributing your video widely — usually because early signals (CTR, early watch time) weren't strong enough to warrant wider distribution.
Click-through rate (CTR) is the percentage of impressions that resulted in a click. Average CTR across YouTube is 2–10%; anything above 4% for a small channel is solid. Very high CTR (above 10%) combined with low retention is a red flag — your thumbnail is over-promising. Low CTR (below 2%) means your thumbnail and title aren't standing out enough in the context where they're being shown.
Average view duration (AVD) is the average number of minutes a viewer watches per view. Absolute AVD depends entirely on video length — 3 minutes of AVD on a 6-minute video is excellent; 3 minutes on a 30-minute video is not. Use percentage watched instead for cross-video comparisons.
Average percentage viewed is AVD as a percentage of video length. Above 50% is generally strong for long-form content. Below 30% consistently suggests either the wrong audience is finding your video, or the video doesn't deliver on what was promised in the hook.
Traffic source metrics
Impressions from search vs. impressions from suggested tells you where your distribution is coming from. Search-driven channels grow more predictably (if you rank for relevant queries) but have lower ceilings. Suggested-driven channels can grow explosively when a video gets picked up — but suggested performance is harder to engineer. Most healthy channels have a mix of both.
Top search queries (visible in the Reach tab under Traffic sources → YouTube search) shows what people actually typed to find your video. This is real keyword intelligence — these are proven search terms your content already ranks for, which means creating more videos on related topics is a lower-risk bet than guessing what to target next.
Audience metrics
Returning vs. new viewers shows you your channel's health in one number. A high proportion of returning viewers means you're building an audience that comes back. A channel that's entirely new viewers is growing but not retaining. If your returning viewer percentage is below 10%, focus on end-screen calls to action and series-style content that creates a reason for viewers to return.
Unique viewers is the estimated number of individual people who watched your content in the selected period. Comparing unique viewers to subscribers tells you what proportion of your subscriber base is actually watching — a useful loyalty metric that views alone can't give you.
Subscriber activity by time (in the Audience tab) shows when your subscribers are most active on YouTube. This is useful for timing uploads — though it matters less than most creators think, because YouTube distributes content over hours and days, not just at publication time. Still, publishing 1–2 hours before your peak activity window is a reasonable default.
Engagement metrics
Likes are a weak signal algorithmically but a useful gut-check. If a video gets significantly fewer likes than your average for similar view counts, the content probably didn't land as well as you hoped. Likes don't directly influence algorithmic distribution, but they correlate with the kind of strong positive reaction that does.
Comments are a stronger engagement signal and the best qualitative data source you have. Read every comment on your early videos — they'll tell you what you got right, what was confusing, and what your audience actually wants to see next. High comment counts on a video, especially thoughtful questions or detailed responses, signal genuine engagement that YouTube does weight positively.
Revenue metrics (post-monetization)
RPM (revenue per mille) is your earnings per 1,000 views, after YouTube's 45% cut. This is the metric that tells you your actual earning rate. RPM varies widely by niche — finance and B2B content can earn $15–$40 RPM while gaming might earn $1–$3 RPM. Once you're monetized, use RPM combined with your view trend to project future earnings.
CPM (cost per mille) is what advertisers pay per 1,000 ad impressions — before YouTube's cut. CPM is higher than RPM by roughly 45%. Knowing both helps you understand whether a CPM change is driving your revenue movement or whether your monetized playback rate has changed.
Try TubeMilestone →