YouTube Analytics for Shorts: the five numbers that matter
For a YouTube Short, five numbers do most of the work: viewed vs swiped away, average view duration, average percentage viewed, the audience retention curve, and subscribers gained. Read them in that order and against your own Shorts of a similar length; views on their own mostly tell you how often the Short was shown.
YouTube Studio gives Shorts their own tab of analytics, and the long-form habits — watch hours, click-through rate on thumbnails — don’t carry over. Here is what each of the five tells you, and what they looked like on one small channel.
What are the five numbers?
| Number | Question it answers | In YouTube’s Analytics API? |
|---|---|---|
| Viewed vs swiped away | Did the first frame stop the scroll? | No, Studio only |
| Average view duration | How many seconds did a view last? | Yes |
| Average percentage viewed | What share of the Short was that? | Yes |
| Audience retention | Where did people leave, or replay? | Yes |
| Subscribers gained | Did anyone want more? | Yes |
Views sit above all of these as the denominator. Since March 2025 YouTube counts a Shorts view each time it starts or replays, and keeps “engaged views” as the older, stricter count, so check which one a chart uses before comparing with an older screenshot.
1. Viewed vs swiped away
The share of people in the Shorts feed who stayed to watch rather than swiping past. It is the Shorts version of Instagram’s skip rate — the first-frame test — though it is counted differently, so compare Shorts with Shorts. It isn’t in the YouTube Analytics API, so read it in Studio.
2 and 3. Average view duration and average percentage viewed
Average view duration is seconds per view; average percentage viewed is that divided by the Short’s length. Read them together: a high percentage on a very short Short can still be only a few seconds watched. Replays count, so the percentage can pass 100%.
What we see in our own data: across the 10 Shorts with YouTube Analytics data on the tomeescapes channel (posted 2026-08-10 to 2026-10-09; data as of 28 October 2026), the median Short was 18.5 seconds long, watched for 13.5 seconds per view, for a median 76% viewed. The middle half sat between 64% and 89%, and 2 passed 100%. Longer Shorts had the lower percentages (correlation -0.54). 10 Shorts is a small sample; it is one channel’s baseline, not a benchmark.
4. Audience retention
The curve shows, second by second, how many views each moment got relative to the number of views. On Shorts it often starts above 100%, because replays watch the start again. A flat line after the first second is good; a step down marks one moment that lost people; a bump marks a moment people went back to. On the same channel, the 10 Shorts’ curves had a median of 122% at 3 seconds and 48% at the last second. The full walkthrough is in YouTube Shorts retention and swiped away.
5. Subscribers gained
Retention says people watched; subscribers say they wanted more of you. It is the slowest number to move and the one closest to what most creators actually want from Shorts.
What we see in our own data: those 10 Shorts had 6,465 views between them and brought 5 subscribers — 4 of the 10 brought any at all. That is about 0.8 per 1,000 views, and it is why we read subscribers across many Shorts rather than one: per Short, the number is mostly zero.
What should you check, in order?
- Viewed vs swiped away against your other Shorts. If more people swiped, fix the first frame before anything else.
- Average view duration and percentage viewed, against Shorts of a similar length.
- The retention curve: where it steps down, and whether it bumps at a loop.
- Subscribers gained, added up over your last ten or twenty Shorts.
- Views last, as context: how widely the Short was shown, not how well it did. For the same question on Instagram, see reach vs views.
How do Shorts numbers compare with Reels?
They don’t, number for number. YouTube counts replays in its retention curve and views; Instagram’s skip rate and graph are built differently. Compare each video with its own platform’s usual, then look at where the two disagree: the same video can hold well on one and not the other. For the Instagram side, start with average watch time on Reels.
Visteria brings in each Short’s views, average view duration, percentage viewed, retention curve and subscribers from YouTube Analytics and shows them against your own Shorts. See what you get.
Questions people ask
What is the most important metric for YouTube Shorts?
There isn’t one. Viewed vs swiped away says whether the opening worked; average percentage viewed and retention say whether people stayed; subscribers gained says whether they wanted more. Read them in that order.
What is a good average percentage viewed for Shorts?
Higher than your own Shorts of a similar length. On one small channel, the median across 10 Shorts was 76%, and longer Shorts scored lower.
Why don’t my Shorts bring subscribers?
Most Shorts bring none; on the channel we measured, 4 of 10 brought any. Read subscribers across many Shorts, and look at whether the ones that did have something in common.