YouTube Analytics Explained: What Each Metric Means

Analytics is the part of YouTube Studio where a channel's own numbers live. Many guides list 15 or more metrics and quote a "good" value for each. YouTube itself gives one numeric benchmark, for click-through rate, and defines the rest without putting a number on what is good. This guide takes the tabs and metrics in YouTube's words, says where the help pages are silent, and adds a table of which numbers anyone can see and what our own data shows about the ones that are public.

Where do you find YouTube Analytics, and what are the tabs?

YouTube Help says to sign in to YouTube Studio and choose Analytics in the left menu for channel reports. For a single video, choose Content, point to the video and select Analytics. Help also notes that from July 2026 YouTube is gradually rolling out an updated Studio, so your page may look different from the descriptions below.

Some data, such as geography, traffic sources or gender, may be limited, Help says. Advanced mode adds an expanded report where you can pick specific data, compare performance and export it.

What do the core metrics mean?

These are YouTube's own definitions, from its Analytics help pages. The wording matters: average view duration, for example, is defined as an average among viewers who stayed to watch.

Two details are easy to miss. Help counts an impression only when the thumbnail is shown for more than 1 second and at least 50% of it is visible. It does not count impressions on external websites and embeds, YouTube Kids, YouTube Music, email or notifications. So the click-through rate describes how a thumbnail does inside YouTube, and it covers only a subset of your total views.

The new, casual and regular groups have exact rules. Help says new viewers watched something on your channel for the first time in the period. Casual viewers watched at least once a month for one to five months in the past year. Regular viewers watched at least once a month for more than six months.

What is a good click-through rate?

The only figure YouTube publishes is that half of all channels and videos have an impressions click-through rate that can range between 2% and 10%. It adds that new videos, channels less than a week old and videos with fewer than 100 views can see an even wider range, and that a video with many impressions, for example on the Home page, naturally has a lower rate. Videos whose impressions come mostly from sources like your channel page may have a higher rate.

Its advice is to compare click-through rates between videos over the long term and to keep the traffic source in mind, because the same thumbnail gets a different rate in search than on the home feed. It also says to avoid clickbait, since YouTube recommends a video when it is relevant to viewers and its average view duration shows they find it interesting. Pages that publish click-through ranges by traffic source or niche rarely say how they measured them, so treat those tables as opinion. Our thumbnail testing guide explains why YouTube picks a test winner by watch time and not by click-through rate.

What do traffic sources and unique reach add?

Two reports change how you read a view count. The content report shows how viewers found your content across browse features, the Shorts feed, suggested videos, YouTube search, channel pages and others, and the same idea drives the click-through rate you saw above, since impressions from different places convert differently. Unique reach is the estimated number of individual people who saw your video or channel. It accounts for shared viewing on devices like YouTube on a TV. Help's example is three people watching a video together on one TV, which counts as one view while unique reach rises by 3. Its companion metric, views (co-viewed), counts a family of three watching twice as six.

How do you read audience retention?

Retention lives on a single video, not on the channel. Help says the key moments for audience retention report is available only at the video level, that the data typically takes 1 to 2 days to process, and that a video needs to be at least 60 seconds long and have at least 100 views for these moments to show.

Help describes four kinds of moment:

• Intro is the percentage of your audience still watching after the first 30 seconds. A high intro percentage can mean the start matched the viewer's expectation from the title and thumbnail, and Help suggests changing the thumbnail and title or the first 30 seconds if it is low.
• Top moments are places where almost no one dropped off. If they fall late in the video, Help suggests introducing the compelling content earlier, because audiences usually shrink over the length of a video.
• Spikes are moments that were rewatched or shared. Help says a spike can mean the audience watched that segment more than the earlier ones, or that the content was unclear and they had to rewatch it.
• Dips are moments people skipped or left. Help suggests reviewing them to understand why interest fell.

A flat line means viewers watched that part from start to finish, gradual declines mean interest fades over time, and Help says videos on YouTube generally taper off during playback. YouTube Help gives no retention benchmark as a number. What it offers instead is a comparison: See more shows how the video compares with all YouTube videos of similar length, typical retention compares your 10 latest videos of similar length, and segments that split retention by new and returning viewers, subscribers and non-subscribers, and organic and paid traffic. The same 10-video comparison is the honest benchmark for average view duration.

How do Shorts metrics differ?

Shorts have their own reports. Help's Shorts key metrics card shows views, likes and subscribers. Its "how many chose to view" figure is the percentage of times viewers viewed a Short against swiping away, and "how viewers engaged" is the percentage of times viewers stayed to watch past the initial seconds. "How viewers find your Shorts" lists sources such as the Shorts feed, search and channel pages.

One point of confusion is what counts. From August 24, 2026 the public view count for a Short and for a video is counted when playback starts. Partner Program earnings still use engaged Shorts views and eligibility uses qualified Shorts views, as Help states. Our Shorts monetization guide explains what those terms mean.

In what order should you read the numbers?

Help does not prescribe a sequence, so this is our suggestion.

1. 01

   Start with Reach on a video

   Check impressions and click-through rate, and the traffic sources behind them. A low rate on Home impressions and a healthy one from search tell different stories.

2. 02

   Move to Engagement

   Compare average view duration and average percentage viewed with your recent videos of similar length, not with a number from a blog.

3. 03

   Open the retention report

   Look at the intro percentage, then the first dip. That is usually where the title, the thumbnail or the opening decides the result.

4. 04

   Check Audience

   See whether views come from new or returning viewers and when your viewers are online, in the report that covers the last 28 days.

5. 05

   Change one thing and wait

   Compare a run of similar videos before and after a change. One video says little, as our guide to why views drop explains.

Our Channel Audit and Video Comparison use the public side of this, which is where the next table starts.

Which YouTube metrics are public and which are private?

The YouTube Data API documentation lists what a public channel or video record contains. Channel statistics carry a view count, a subscriber count that the docs say is rounded to three significant figures, a flag for a hidden subscriber count, and a video count. Video statistics carry a view count, like count and comment count. The API also returns the publish time and the duration. The metrics that describe an audience's behavior sit in a separate API, the YouTube Analytics API, whose documentation describes reports for channel owners and content owners.

That split decides what any outside tool can honestly compute. A tool that reads public data can work out views per day, likes per view, comments per view and upload cadence. It cannot know a channel's click-through rate, watch time or retention, and a site that shows them for someone else's channel is estimating. Our tools stay on the public side and label what they cannot see.

What does public data show? Likes and comments per view

Public engagement rates are the closest outsiders get to how a video landed, so we measured them. We read the statistics of the latest 50 uploads of 115 channels drawn at random from the ones our site tracks (median 274,000 subscribers, so the set leans toward larger channels), through the Data API on September 25, 2026. We kept videos at least 14 days old with at least 1,000 views. We called a video long-form when it ran longer than 3 minutes, and a Short when it was 3 minutes or less and YouTube served it at its youtube.com/shorts/ address, which is behavior we observed and not a documented feature.

Likes were hidden on 28 of the 1,630 long-form videos and 67 of the 1,707 Shorts, and comments were switched off on 20 Shorts, and those are left out of the matching rate. The spread is wide: the middle half of long-form videos runs from about 1 to 3.7 likes per 100 views. A Short draws more likes per view than a long video but far fewer comments per view.

For long-form videos the rate also moved with how many views a video had.

In our sample the rate fell as views rose, and we did not test why. Read that as a reason not to hold a small video and a hit to the same like rate. It is not a rule. The definition of a view also changed on August 24, 2026, so rates measured across that date are not directly comparable. Our channel audit guide and video comparison guide show how to use these public rates side by side.

FAQ

Related reading