The metrics that matter

0:00 The metrics that matter

Open YouTube Studio and you'll find dozens of numbers. Impressions, click-through rate, average view duration, RPM, engaged views, unique viewers. It's easy to either ignore them all, or panic about every one. The truth is that a handful of metrics answer three simple questions: are people choosing my videos, are they watching, and are they coming back? In this lesson, you'll learn the core metrics and what they really mean, how to read click-through rate and retention in context, the difference between RPM and CPM, the Shorts views change, and how to build a simple dashboard.

0:42 Three questions

Here's the key idea: analytics are feedback from your audience, organized into three questions. Are people choosing my videos? That's impressions and impressions click-through rate. Are they watching? That's average view duration, average percentage viewed, audience retention and watch time. Are they coming back? That's returning and new viewers, unique viewers and subscribers gained per video. Then there's engagement, likes, comments and shares, and for Partner Program channels, revenue metrics like estimated revenue and RPM.

1:15 Definitions

Let's get the definitions right. Impressions are the times your thumbnails were shown on YouTube, on certain surfaces, not including external traffic. Click-through rate is the share of those impressions that became views. Average view duration is the average time watched per view, and it depends heavily on video length. Average percentage viewed is that duration as a share of video length, so longer videos usually have lower percentages. Unique viewers are an estimate of distinct people, which is a better indicator of audience size than subscribers. And watch time is total hours watched, important for reach and for Partner Program eligibility on long-form.

2:00 CTR in context

Now click-through rate in context, because it's the most misread number on YouTube. There's no universal good click-through rate. It varies by niche, traffic source, thumbnail style and audience. And here's the pattern that surprises people. CTR is usually highest when a video is first shown to your core audience, and it declines as YouTube shows it to broader audiences who know you less. So a falling click-through rate alongside rising impressions can be a sign of success, not failure. Think of it like a shop that gets featured in a city guide. More people walk past, and a smaller share come in, but the total number of customers goes up.

2:48 Retention + Shorts views

Retention also needs context. A three-minute tutorial and a forty-minute documentary aren't comparable. Use YouTube's comparison with typical videos of similar length, and your own averages for similar videos. And now the Shorts change. Since March thirty-first, 2025, a Shorts view counts whenever a Short starts to play or replay. The older, stricter measure is now called engaged views, available in Advanced mode, and it's what YouTube uses for Partner Program eligibility and Shorts revenue. So when you compare Shorts over time, use engaged views and viewed versus swiped away, not raw views.

3:28 RPM vs CPM

RPM versus CPM, which many creators confuse. CPM, cost per mille, is how much advertisers pay per thousand ad impressions, before YouTube's share. RPM, revenue per mille, is what you actually earn per thousand views from all revenue sources, after YouTube's share. RPM varies widely with audience geography, niche, season and format. Advertisers often spend more in certain periods, like the shopping season at the end of the year. So never compare your RPM with another creator's without considering who their audience is and where.

4:05 Example 1: first week of a tutorial

A simple example, with illustrative numbers. A tech tutorial's first week: eighty thousand impressions, a click-through rate of six point one percent, where the channel's typical range for tutorials is five to seven. Fifty-three hundred views. Average view duration of four minutes ten seconds on a nine-minute video, forty-six percent viewed, versus the channel's typical forty. Traffic: forty-five percent search, thirty percent suggested. Returning viewers, thirty-five percent. Interpretation: packaging is normal, retention is above typical, and search is strong, so the video will likely keep earning evergreen views. Action: make a follow-up on the next most-searched related question, and link them.

4:49 Example 2: UK Urdu cooking

Now a realistic scenario, illustrative. Maryam runs a UK-based Urdu cooking channel. Her subscriber count is stuck, and she's discouraged. But when she looks at unique viewers and returning viewers instead, she sees that unique viewers have doubled this year, and returning viewers are steady. Her audience is growing, it just isn't clicking subscribe. Her RPM is also higher than a friend's channel, because more of her viewers are in the UK. She stops obsessing over the subscriber counter, adds a spoken subscribe reminder only after delivering the recipe, and focuses on series that bring people back.

5:31 Watch me: simple dashboard

Watch me build the simple channel dashboard from the lesson. In YouTube Studio's Advanced mode, I choose the table by video for the last ninety days, and export it to Google Sheets. I add four calculated columns. CTR versus typical: the video's click-through rate divided by my channel's median for the same format. Average view duration percent: average view duration divided by video length. Subscribers per thousand views. And returning share: returning viewers divided by returning plus new. Then I make one chart per column, by video and by pillar. It takes fifteen minutes a month, and it's the one table I actually look at.

6:17 Traffic + audience reports

Traffic sources and audience reports complete the picture. Traffic sources tell you how viewers found each video: browse, suggested, search, Shorts feed, external, playlists or notifications. YouTube search terms show which queries led to views. Suggested video sources show which videos yours was recommended alongside. And the audience tab shows geography, age, devices, when your viewers are on YouTube, and, where available, other channels and videos your audience watches. The common mistakes: comparing CTR or retention across very different formats, panicking over first-hour numbers, obsessing over subscribers, confusing CPM with RPM, and ignoring traffic sources, which explain most performance differences.

7:00 Weekly analytics routine

Here's a weekly analytics routine that takes twenty to thirty minutes. Check the channel overview for trends compared with the previous period. Review your last upload against its typical performance at the same age. Look at the top traffic sources for your recent videos. And read the most recent comments. Then write down one insight and one action. That's it. A short, consistent routine is far more useful than an occasional deep dive that never changes what you make. If you work with a team, share the one insight and one action in a short message every week, so everyone learns together.

7:44 Recap

Recap. Analytics answer three questions: are people choosing, watching and returning? CTR naturally falls as impressions broaden, so compare to your own typical range. Compare retention only across similar formats and lengths. For Shorts, use engaged views. RPM is what you earn per thousand views; CPM is what advertisers pay. And keep a simple dashboard you actually look at.

8:10 Try this now

Try this now. Pull the core metrics for your last five videos into a table: impressions, CTR, average view duration percent, main traffic source and returning viewers. Compare each with your channel's typical range for similar videos. Write one insight per video. Then set up the monthly export and the four calculated columns. Next lesson: diagnosing why a video under- or over-performed.