What Is a Good Retention Rate for YouTube Shorts? 2026 Benchmarks

YouTube publishes no official retention benchmarks for Shorts, so every “good” number is a rule of thumb. The ranges creators most often report in 2026: a Stayed to watch rate of 60-79% is healthy (70% is the usual target), 80% or more is breakout territory, and under 60% usually means the opening is losing people. For average percentage viewed, healthy is roughly 80-90% for Shorts under 35 seconds and 60-75% for Shorts over a minute. On our own small test channels, two AI story Shorts held 69.9% and 60.7% average percentage viewed, while tutorial-style Shorts kept only 16-41% of viewers past the first seconds and stalled.

This guide is for Shorts creators, faceless channel operators and marketers who open YouTube Studio, see a percentage and want to know whether it's good. It defines each retention metric exactly as YouTube Studio names it, shows where to find it, gives benchmark tiers by Short length, uses our own numbers as a worked example, and ends with a routine for fixing weak Shorts. Figures are current as of October 2026.

Related guides: engagement rate benchmarks for TikTok, Reels and Shorts, how to improve retention on faceless videos and does YouTube suppress AI-generated videos.

What retention metrics does YouTube Studio show for Shorts?

YouTube Studio reports Shorts retention through Stayed to watch, average view duration, average percentage viewed and the audience retention curve, with views and engaged views as the counts underneath. The definitions below are quoted from YouTube's Help Center article on content performance (YouTube Help); the names are the ones Studio shows in October 2026.

• Stayed to watch: “the percentage of times viewers stayed to watch past the initial seconds of a Short.” Its complement, shown beside it, is Swiped away. Older guides call the pair “viewed vs. swiped away.”
• Views: since March 31, 2025, a Shorts view counts every time a Short starts to play or replays, with no minimum watch time.
• Engaged views: the older count, “how many times viewers stayed to watch past the initial seconds, not including any loops,” kept in YouTube Analytics.
• Average view duration (AVD): the average time watched “among those who stayed to watch,” calculated from engaged views and their watch time.
• Average percentage viewed (APV): the average share of the Short watched among those who stayed to watch.
• Audience retention: the curve in a Short's analytics showing what share of viewers is still watching at each moment; on looping Shorts it can rise above 100%.

The key detail is in the small print. Average view duration and average percentage viewed are calculated only over the viewers who stayed past the first seconds. A Short can show a respectable 70% average percentage viewed while most of the people who saw its first frame swiped away. Read the two numbers together, never one alone.

The 2025 view-counting change matters when you compare Shorts over time. Because every start and replay now counts as a view, a view count says less about attention than it did before March 2025, so compare older and newer Shorts on engaged views. YouTube's Help Center says the Partner Program still pays and qualifies channels on engaged and qualified views rather than the new count (YouTube Help), and from August 24, 2026 the same start-to-play rule applies to long-form videos and live streams too.

Where do you find Shorts retention in YouTube Studio?

Go to Analytics, open the Content tab and choose the Shorts chip; the “How viewers engaged” card shows Stayed to watch across your recent Shorts. For a single Short, open its analytics from your content list, where you'll find Stayed to watch, average view duration, average percentage viewed and the retention curve, plus the traffic sources that show how much of its audience came from the Shorts feed, which YouTube defines as traffic from the Shorts vertical viewing experience (YouTube Help).

Two habits make the numbers more useful. Wait at least 48 hours after posting, because a Short's first test audience is small and early percentages swing. And compare like with like: a 25-second loop and a 90-second story are different products, so benchmark each against Shorts of similar length on your own channel. The most careful third-party guides give the same advice.

What is a good Stayed to watch rate for YouTube Shorts?

As a rule of thumb, 60-79% is healthy, 80% or more is breakout territory, and under 60% means the opening is costing you the audience. YouTube publishes no target. 1of10, a YouTube analytics company, describes 70-79% as the community-observed “working” range and 80% and above as strong, and says plainly that these figures come from creator-reported data, not from YouTube (1of10). We set the healthy floor at 60% rather than 70% because the creator guides we reviewed disagree about where “average” ends, and a Short at 60-69% is one to improve, not one to write off; treat 70% as the target and 60% as the floor.

Our data adds a caution for small channels. The tutorial Shorts on our busiest test channel held only about 30% Stayed to watch, yet 10 of 12 of them passed 600 views in late September (the best reached 1,588). Later uploads on that channel fell to 23.4% and 15.9%, and distribution stopped. A low rate doesn't mean zero views on a new channel; it means YouTube runs out of reasons to keep testing. Struggling Shorts can get a burst, but they rarely compound.

Almost everything that moves this number happens in the first one to three seconds: the first frame, the first spoken line and the first on-screen words. Our video hooks guide has templates, and the free YouTube hooks generator drafts openers you can test.

What is a good average percentage viewed for Shorts?

It depends on length: roughly 80-90% is healthy for Shorts under 35 seconds, about 65-80% for Shorts of 35 to 60 seconds, and 60-75% for Shorts over a minute. The under-35 and over-60 ranges are 1of10's community-observed working ranges; the middle band is our interpolation between them, because no source we found publishes one.

For a reference point from a large, established channel, vidIQ reports that its own Shorts averaged 73.6% average percentage viewed over the 12 months to June 2026 (vidIQ). The same post puts a good figure for 6- to 9-minute long-form videos at roughly 30%, which is why Shorts and long-form retention should never be compared directly.

Loops complicate the top end. Very short Shorts that viewers watch twice can report an average percentage viewed above 100%, which creators treat as the strongest signal a Short can send. If your best numbers all come from sub-20-second loops, that is a format insight, not proof that every Short should be that short.

What is a good average view duration for a YouTube Short?

A good average view duration is the healthy percentage multiplied by the Short's length: about 24-27 seconds on a 30-second Short, 29-36 seconds on a 45-second Short, 39-47 seconds on a 60-second Short and 54-67 seconds on a 90-second Short. Seconds are the number to watch when you're choosing a length, because a longer Short that holds a lower percentage can still keep each viewer longer.

Worked example: a 30-second Short at 85% holds the average engaged viewer for about 25.5 seconds; a 60-second Short at 65% holds them for 39 seconds. The longer one keeps people longer even though its percentage is lower. Short, looping formats make high percentages easier, while a story that needs 60 seconds to land can be worth the lower figure. The free video length calculator converts a script's word count into seconds, which helps you choose a length before you record.

YouTube Shorts retention benchmarks at a glance (2026)

The table combines the tiers above. Treat every row as a rule of thumb for judging your own Shorts, not a YouTube threshold, and compare against your channel's own history first.

Two cautions. These tiers assume most of a Short's views came from the Shorts feed; a Short that is found mostly through search or your channel page behaves differently. And tiers are for diagnosis, not judgment: a Short can be struggling on Stayed to watch and still be your channel's best performer if everything else you post is weaker.

What does good look like? Our own Shorts, with numbers

Here is our own data, offered as a worked example rather than a benchmark. Between September 16 and October 5, 2026, we published 38 Shorts on seven test channels, all under 200 subscribers, and read retention in YouTube Studio for seven of them on four channels. Every Short was built around AI-generated narration and visuals. Two formats were in the mix: story Shorts posted exactly as Vidpal Studio generated them, and tutorial Shorts that opened by scrolling a famous channel's page before revealing a Studio result.

The best story Short is what healthy looks like. The hip hop history Short ran 56 seconds and the average engaged viewer watched 39 of them, 69.9%, inside our 65-79% healthy band for 35 to 60 seconds. It reached 1,209 views on a channel with four subscribers, 97.9% of them from the Shorts feed, and its views flattened after about eight hours. The Stoic story held 60.7% on only 89 views, and a procrastination explainer on the same channel held 39.0%, which shows that format alone doesn't guarantee retention; the topic and the first line still matter.

The tutorials show what struggling looks like, and why it is easy to misread. On the busiest channel they still collected views, because a 30% Stayed to watch rate on a fresh channel earned tests of several hundred to about 1,600 views. But the 40-second tutorial on another channel kept only 41.4% of viewers past the opening, and those who stayed watched 27% of it, about 11 seconds. Multiply the two and the average person who saw its first frame watched roughly 11% of the video. Later tutorials on the bigger channel fell to 23.4% and 15.9%, and its four Shorts from October 1 to 4 drew 64, 162, 2 and 6 views.

The lesson we took: the AI output itself could hold viewers, and the wrapper we put around it lost them in the first seconds. It is a common Shorts retention problem, and it shows up in Stayed to watch long before it shows up in views. One more caveat from the same review: the channel with the most views was shown mostly to viewers aged 13 to 17, who watched but rarely subscribed (8 new subscribers from about 11,000 views). Retention tells you whether viewers watched, not whether they were the viewers you wanted.

Why can Stayed to watch and average percentage viewed disagree?

Because they measure different groups of people. Stayed to watch is the share of plays where the viewer got past the opening; average view duration and average percentage viewed only count the people who did. A Short with a weak opening and a strong middle can show 45% Stayed to watch and 85% average percentage viewed: most people left at once, and the few who stayed watched almost all of it. Each combination points to a different fix:

• Low Stayed to watch, high average percentage viewed: the opening is the problem and the rest works. Rework the first frame and the first line.
• High Stayed to watch, low average percentage viewed: the hook works but the middle sags. Cut the slowest section, shorten the Short or promise a payoff up front.
• Both low: the idea or the packaging doesn't match what viewers expected. Change the idea, not the edit.
• Both high, few views: the Short is good and the test audience was small. Keep the format and judge it over five to ten uploads, not one.

To compare Shorts fairly, multiply the two. Stayed to watch times average percentage viewed approximates how much of the video the average viewer saw, including the ones who swiped. A Short at 70% and 75% scores about 0.53; one at 45% and 85% scores about 0.38, even though its average percentage viewed looks better on its own.

How does Shorts length change what counts as good retention?

Longer Shorts need lower percentages to count as healthy, but they need a script that earns every extra second. Shorts can run up to three minutes (YouTube Help), and at a typical narration pace of about 150 words per minute, a 30-second Short holds roughly 75 words, a 60-second Short about 150 and a 90-second Short about 225.

The trade-off is simple. Short scripts make high percentages and loops easier. Long scripts make room for a story with a payoff but give viewers more places to leave. Pick the length the idea needs, then cut anything that doesn't move it forward. Our guide to YouTube Shorts length and size requirements covers the specs, and ideal video length for every platform compares Shorts with TikTok and Reels.

In our sample, the best average percentage viewed came from a 56-second story, not from the shortest Short, so don't cut a video that has a real arc just to chase a percentage.

How to improve Shorts retention: a 7-step routine

Step 1: put your last 10 Shorts in a sheet with their length, Stayed to watch, average percentage viewed, average view duration and main traffic source. Record each one at least 48 hours after posting.

Step 2: sort by Stayed to watch and compare your top three openings with your bottom three. Write down what happens in the first two seconds of each.

Step 3: fix the opening first. Show the result, the question or the most surprising fact in the first frame, and cut logos, title cards and slow pans.

Step 4: then fix the middle. Open the retention curve, find the steepest drop after the opening and cut or tighten that section.

Step 5: match length to content. If average percentage viewed sits under the healthy band for its length, test the same idea 15 seconds shorter.

Step 6: make captions readable: large, high-contrast, a few words at a time, kept clear of the buttons on the right and the title area at the bottom.

Step 7: change one variable per batch of five Shorts and judge the batch, not single uploads, because single Shorts swing widely.

Our retention guide for faceless videos goes deeper on pacing and structure, and how to get more views on YouTube Shorts covers what to do once retention is in the healthy band.

Common mistakes when reading Shorts retention

• Judging a Short by views alone. Since March 2025 every start and replay counts as a view, so views can rise while attention falls.
• Reading average percentage viewed without Stayed to watch. It covers only the people who stayed, so it can hide a broken opening.
• Comparing Shorts with long-form. A 30% average percentage viewed is good for an 8-minute video and poor for a Short.
• Comparing different lengths directly. Benchmark a 90-second story against your other long Shorts, not against 20-second loops.
• Checking too early. Percentages in the first hours come from a tiny test audience.
• Copying someone else's benchmark as a target. Niche, length and audience shift every number, and your own last 20 Shorts are the best baseline.

What can't these benchmarks tell you?

They can't predict views. YouTube publishes no Shorts retention benchmarks and doesn't say which level triggers a wider test. The tiers here combine creator-reported ranges (1of10's are not audited), one large channel's published average (vidIQ's) and our own seven-Short sample from four small channels over three weeks. That is enough to tell a struggling Short from a healthy one, and not enough to forecast reach.

Retention is also only one input. Topic demand, the audience a Short reaches and how distinct your channel is all affect distribution, and YouTube's July 2026 policy clarification is a reminder that templated, repetitive channels can lose monetization whatever their retention. Our post on whether YouTube suppresses AI-generated videos, linked at the top, covers that side.

Studio levers for retention, and which Vidpal plan fits

Vidpal Studio gives you direct control over the three levers this post keeps returning to: the opening, the captions and the length. Retention itself is measured in YouTube Studio, not in Vidpal, so the reading and comparing in this post stays with you.

The opening: Studio's Hook Intro puts something in front of the narration's first line. Choose a Viral Clip from the curated library, generate one with Text → Video (portrait or landscape) or Image → Video (a 5-second 9:16 clip, limited to 5 per hour), or use a Text Reveal. A generated hook clip costs 1 extra credit, charged only when it succeeds. If your Stayed to watch is under 60%, this is the first thing to test.

The captions: 75 caption presets, such as Mozi, Beasty, Karaoke, Hook Punch and Highlighter Box, apply to the whole video, with about 18 animations including Karaoke Pop and Word Punch and a Lines control for one word or three at a time. Save your choice in a Brand Kit so every Short matches.

The length: pick 30, 45, 60, 75 or 90 seconds and Studio calibrates the script to about 150 words per minute. If a section drags, rewrite it with AI, regenerate its image or swap in stock footage; section image prompts can't be edited directly.

Which plan fits depends on how many Shorts you want to test. Each Studio Short costs 1 credit and the first export is free; a re-render after changes costs again.

Starter fits a test of three or four Shorts a week. Pro covers a daily Short with room to try hook clips on about a third of them, and adds publishing and scheduling to YouTube, Instagram, TikTok and LinkedIn through the Social Scheduler. Business covers a daily Short with a hook clip on every one and leaves credits for long videos or re-renders. Annual plans grant the year's credits upfront; see pricing for details.

Where Studio isn't the right tool: if your Shorts are clips of you talking to camera, Vidpal's AI Editing (silence removal, auto zooms, captions and Smart Reframe for landscape footage) fits better, and if you need an AI avatar presenter, Vidpal doesn't make those. Start free on Vidpal to build a Short and try the hook and caption options; exporting Studio videos starts on Starter.

Frequently Asked Questions

What is a good retention rate for YouTube Shorts? There is no official figure, but the common rules of thumb in 2026 are a Stayed to watch rate of 60-79%, with 70% the usual target and 80% or more for breakout Shorts, plus an average percentage viewed of about 80-90% for Shorts under 35 seconds or 60-75% for Shorts over a minute.

What does Stayed to watch mean in YouTube Studio? It is the percentage of times viewers stayed to watch past the initial seconds of a Short instead of swiping away, as defined in YouTube's Help Center. Its complement is Swiped away, and older guides call the pair viewed versus swiped away. It is the clearest measure of whether your opening works.

What is a good average view duration for a 60-second Short? About 39 to 47 seconds, which is 65-79% of the video, is a healthy range by the rules of thumb in this guide. YouTube calculates average view duration only among viewers who stayed past the first seconds, so read it alongside Stayed to watch before deciding a Short worked.

Can average percentage viewed be over 100% on YouTube Shorts? It can, according to creator-reported data. When viewers loop a Short and watch it more than once, the average share watched can exceed the video's length, which happens most on very short Shorts. Treat it as a strong signal, but compare it only with other Shorts of similar length.

Why do my Shorts have high average percentage viewed but low views? Usually because few people stayed past the opening. Average percentage viewed is calculated only among viewers who stayed to watch, so a Short can hold 85% of a small group while most viewers swiped away. Check Stayed to watch first, and if it is under about 60%, rework the first one to three seconds.

Did the 2025 change to Shorts view counting affect retention metrics? It changed what a view means. Since March 31, 2025, every start or replay counts as a Shorts view with no minimum watch time, while the older measure continues as engaged views. Average view duration is calculated from engaged views, so compare older and newer Shorts on engaged views rather than on views.

How many Shorts should I post before judging retention? Judge formats over batches of five to ten Shorts, not single uploads. Single Shorts swing widely, especially on small channels where each test audience is tiny. Wait 48 hours before reading a Short's numbers, then compare the batch's typical Stayed to watch and average percentage viewed with your previous batch.