Skip to content

Video Retention Benchmark Checker

Each platform is judged against its own benchmarks. Shorts and TikTok videos are never compared with long-form YouTube videos.

Whole minutes, 0 to 720.

Whole seconds, 0 to 59. Shorts are 3 minutes or shorter; TikTok videos are 10 minutes or shorter.

For YouTube, enter average percentage viewed from YouTube Studio. For TikTok, enter the completion rate from TikTok Analytics. For Shorts, average percentage viewed can pass 100% when viewers replay.

Your retention rating

How the rating was worked out

      This checker rates retention against published creator-economy benchmarks, verified 21 November 2026. Neither YouTube nor TikTok publishes official retention targets, and no figure here is a platform-official number. Benchmarks are length bands for the whole platform: they do not adjust for niche, audience, or traffic source, and your own trailing average over your last five videos of similar length is the fairer comparison. Platforms change their metrics and counting rules without notice, so check the result against your own analytics before you rely on it.
      A good retention rate depends on the video length. The platform average for YouTube is 23.7%, and 50 to 70% is strong for videos under 5 minutes. For Shorts, 70% average percentage viewed or above is healthy, and above 100% means viewers replay. For TikTok, the average completion rate is 38%, with top-quartile creators at 55% or more. Never compare retention across lengths: enter your video above to see its rating against the band for its own length.

      Last reviewed: 21 November 2026. Reviewed every 180 days; next review due 19 May 2027.

      How to use the Video Retention Benchmark Checker

      This checker answers one question about one video: is the retention you are seeing good, average, or weak for a video of that length on that platform. It takes the video length and the retention percentage from your analytics, finds the length band the video belongs to, and rates the number against the published benchmarks for that band. The rating is always length-matched, because a retention percentage means nothing without the length behind it. A 40% retention rate is strong on a 25-minute video and weak on a 3-minute video, and the checker says so instead of handing you a single pass-or-fail number.

      To check a video, follow these steps:

      1. Pick the platform. YouTube video (long-form), YouTube Short, or TikTok video. Each platform has its own benchmark tables, and Shorts and TikTok videos are never judged against long-form YouTube numbers. A Short at 70% and a long-form video at 70% are different achievements, and the checker keeps them apart.
      2. Enter the video length in minutes and seconds. Whole numbers only: minutes 0 to 720, seconds 0 to 59. An 8-minute video is 8 minutes and 0 seconds. A 45-second Short is 0 minutes and 45 seconds. The length decides the band, and the band decides the thresholds, so enter the length exactly as it appears in your analytics.
      3. Enter the observed retention percentage. For YouTube, use average percentage viewed from YouTube Studio. For TikTok, use the completion rate from TikTok Analytics. For Shorts, use average percentage viewed, which can pass 100% when viewers replay: the checker accepts up to 200% for Shorts, and up to 100% for long-form and TikTok, where the number cannot go above 100%.
      4. Read the rating. There are four: Below average, Average, Strong, and Exceptional. The rating comes with the band it was judged against, the average view duration in clock form, the gap to the next tier in percentage points, and the comparison with the platform reference figure.
      5. Read the warnings. Every result carries the cross-length warning, because it is the single most common way creators misread their own numbers. Some results carry extra warnings: TikTok videos under 15 seconds and over 3 minutes, and Shorts whose length changes what counts as healthy.
      6. Read the working. The steps show the arithmetic in full: the length in seconds, the band, the thresholds, the average view duration, and the reference comparison. Every number in the steps is the number the rating was made from, so you can check the reasoning instead of trusting it.

      Try the example buttons under the form. The instructive pair is the 8-minute YouTube video at 42% and the 25-minute YouTube video at 28%. The first is Strong and the second is Average, even though 28% is the lower number, because the 25-minute video is judged against a lower band. The 3-minute video at 62% is Strong too, and the 30-second TikTok at 50% is Strong on TikTok's own table. Four different numbers, four different bands, and the same honest question answered each time: good for what length.

      One thing the checker does not do: it does not read your analytics. You type the numbers in from YouTube Studio or TikTok Analytics, and the checker rates what you typed. That keeps the tool fast and private, and it means the rating is only as good as the number you enter. The section on where to find your numbers tells you exactly which figure to copy from each platform, so the number you enter is the one the benchmarks were built for.

      How it works

      The checker runs four steps. First it turns your minutes and seconds into a total length in seconds. Then it finds the length band for that total on the platform you picked. Then it compares your retention percentage with the three thresholds of that band: the average floor, the strong floor, and the exceptional floor. Then it works out the average view duration, the gap to the next tier, and the comparison with the reference figure for the platform. The rating is the tier your number lands in. Nothing else enters into it: no view count, no subscriber count, no niche adjustment. The benchmarks are length bands for the whole platform, and the checker applies them as published.

      The bands and thresholds are the heart of the tool, and each one is attributed to its source in the tables below. For YouTube long-form, the average floors and exceptional floors come from Alan Spicer's audit table, built from hundreds of channel audits and data he accessed during his time on vidIQ's Creator Success team. The strong floors come from Retention Rabbit's 2025 benchmark report, covering more than 10,000 videos across more than 75 niches, as published by TubeLab. The platform average of 23.7% is Retention Rabbit's figure. For Shorts, the thresholds come from Humble and Brag's 2026 Shorts benchmarks, with vidIQ's own Shorts average of 73.6% as the reference. For TikTok, the thresholds come from Viralo's 2026 completion-rate table, with Socialinsider's 2026 average completion rate of 38% as the reference. Neither YouTube nor TikTok publishes official retention targets, and this checker never presents any benchmark as a platform-official figure.

      The worked examples below show the arithmetic on real inputs. Every number recomputes exactly, and each example is also one of the test cases the tool is verified against.

      Worked example 1: 8-minute YouTube video at 42%

      Length: 8 minutes 0 seconds = 480 seconds. Band: 5 to 10 minutes, which covers 300 to 599 seconds. Thresholds for that band: average starts at 35%, strong at 40%, exceptional at 65%. The result of 42% sits at or above the strong floor and below the exceptional floor, so the rating is Strong. Average view duration: 42% of 480 seconds = 201.6 seconds, which is 3 minutes 22 seconds. Against the reference: 42% is 18.3 points above the platform average of 23.7%. Gap to the next tier: Exceptional starts at 65%, which is 23 points above this result.

      Worked example 2: 25-minute YouTube video at 28%

      Length: 25 minutes 0 seconds = 1,500 seconds. Band: 20 minutes or longer, which covers 1,200 seconds or more. Thresholds for that band: average starts at 25%, strong at 40%, exceptional at 55%. The result of 28% sits at or above the average floor and below the strong floor, so the rating is Average. Average view duration: 28% of 1,500 seconds = 420 seconds, which is 7 minutes exactly. Against the reference: 28% is 4.3 points above the platform average of 23.7%. Gap to the next tier: Strong starts at 40%, which is 12 points above this result. Notice that 28% earns Average here while it would be Below average on an 8-minute video, where the average floor is 35%. The length changes the verdict, which is the whole point of the tool.

      Worked example 3: 40-second Short at 85%

      Length: 0 minutes 40 seconds = 40 seconds. Band: YouTube Shorts, a single band covering 180 seconds or shorter. Thresholds: average starts at 50%, strong at 70%, exceptional at 100%. The result of 85% sits at or above the strong floor and below the exceptional floor, so the rating is Strong. Average view duration: 85% of 40 seconds = 34 seconds. Against the reference: 85% is 11.4 points above the 73.6% that vidIQ's own Shorts averaged over the 12 months to June 2026. Gap to the next tier: Exceptional starts at 100%, which is 15 points above this result. Because the Short is 40 seconds long, the under-35-second warning does not apply: that warning fires only for Shorts under 35 seconds rated Strong below 80%.

      Worked example 4: 30-second TikTok at 50%

      Length: 0 minutes 30 seconds = 30 seconds. Band: 30 to 60 seconds, which covers 30 to 59 seconds. Thresholds for that band: average starts at 30%, strong at 45%, exceptional at 60%. The result of 50% sits at or above the strong floor and below the exceptional floor, so the rating is Strong. Average view duration: 50% of 30 seconds = 15 seconds. Against the reference: 50% is 12 points above the average TikTok completion rate of 38% from Socialinsider's 2026 analysis. Gap to the next tier: Exceptional starts at 60%, which is 10 points above this result.

      Two details of the arithmetic are worth knowing. Retention is rounded to 2 decimal places at parse time, after the bound checks and before any computation, so a value typed as 42.005 becomes 42.01% and is rated as 42.01%. And the rating is always made on the rounded value, while the average view duration is rounded to 1 decimal place in seconds and then shown in clock form rounded to the nearest second. The steps print both the seconds and the clock form, so you can see when the clock form is approximate: 201.6 seconds shows as about 3:22, while exactly 420 seconds shows as 7:00 with no qualifier.

      YouTube long-form benchmarks

      Long-form YouTube retention is measured as average percentage viewed: the average share of the video that viewers watched. YouTube Studio reports it alongside average view duration, and the two numbers are the same fact in different units. A 10-minute video with a 4-minute average view duration has 40% average percentage viewed. The benchmarks below are the thresholds this checker rates against, and the table shows where each threshold comes from. All figures were verified on 21 November 2026.

      Video lengthBelow averageAverageStrongExceptional
      Under 5 minutesUnder 40%40% to 49%50% to 69%70% or more
      5 to 10 minutesUnder 35%35% to 39%40% to 64%65% or more
      10 to 20 minutesUnder 30%30% to 34%35% to 59%60% or more
      20 minutes or longerUnder 25%25% to 39%40% to 54%55% or more

      The average floors and exceptional floors are Alan Spicer's, from the audit table he uses when reviewing channels, built on hundreds of audits and data he accessed during his time on vidIQ's Creator Success team. The strong floors are Retention Rabbit's, from the 2025 benchmark report covering more than 10,000 videos across more than 75 niches: 50 to 70% is healthy for videos under 5 minutes, 40 to 55% is solid for 5 to 10 minutes, 35 to 45% is good for 10 to 20 minutes, and 25 to 35% is respectable for videos over 20 minutes. The two sources agree closely on the shape of the ladder, and they disagree in exactly one place: the strong floor for videos of 20 minutes or longer. Retention Rabbit calls 25 to 35% good for that length, while Spicer puts the average band at 25 to 40% and strong at 40 to 55%. This checker follows Spicer for that band, because Retention Rabbit's 25 to 35% good range overlaps its own average band and cannot anchor a four-tier ladder, while Spicer's table gives a clean floor for every tier. The disagreement is recorded here rather than hidden, because a creator with 30% on a 25-minute video should know that one respected source calls that good and another calls it average.

      The platform average of 23.7% is Retention Rabbit's figure for the average YouTube video. It sits below every band's average floor, which surprises most creators the first time they see it, and it is the reason the checker shows the reference comparison on every result. A video rated Average is already beating the platform average by a wide margin. Retention Rabbit also found that more than 55% of viewers leave within the first minute regardless of video length, and that videos in the 5 to 10 minute range hold viewers best, at 31.5% average retention. That last figure is worth pausing on: the length band that holds viewers best still averages under a third of the video watched. The bands in the table are targets to aim for, not descriptions of what most videos do.

      Retention also varies by niche, and the niche can matter as much as the length. Retention Rabbit's report puts educational how-to content at 42.1% average retention, the highest of any category it measured, and vlogs at 21.5%, one of the lowest. A vlogger comparing a 25% result with a how-to creator's 45% is comparing two different audiences with different reasons to stay. The checker does not adjust for niche, because no published benchmark table breaks the tiers down by niche and length together, but the guidance on every result points to the fairest comparison available: your own last five videos of similar length, which YouTube Studio shows as the Typical retention range on the retention graph.

      One more YouTube figure belongs in this section because it governs the first 30 seconds, where most of the damage happens. vidIQ puts 60% of viewers still watching at the 30-second mark as the target for a good hook, with the strongest hooks holding over 65%. YouTube Studio rates the intro on its own as typical, above, or below, and the intro number is the single highest-leverage figure on the whole report: retention rarely recovers what it loses in the first 30 seconds. The checker rates the whole-video percentage, not the intro, but the fix sentence on a below-average result points at the opening first, because that is where the loss usually sits.

      YouTube Shorts benchmarks

      Shorts retention is measured as average percentage viewed, the same metric name as long-form, but it behaves differently. Shorts loop automatically and last seconds, so the percentages run far higher than long-form, and the number can pass 100% when viewers watch the Short more than once. vidIQ reports that its own Shorts averaged 73.6% average percentage viewed over the 12 months to June 2026, against 30.3% on its long-form videos. That gap is why the checker never compares a Short with a long-form video, and why the Shorts thresholds sit where they do.

      RatingAverage percentage viewedWhat it means
      Below averageUnder 50%A structural problem: the hook, the pacing, or the promise is failing
      Average50% to 69%Watchable, but below the healthy bar
      Strong70% to 99%Healthy: the algorithm keeps serving it to new audiences
      Exceptional100% or moreViewers watch the whole Short, and above 100% they replay it

      The thresholds come from Humble and Brag's 2026 Shorts benchmarks: a healthy average percentage viewed for Shorts is 70% or above, below 50% indicates a structural problem, and above 100% means viewers are replaying the Short, which is one of the strongest positive signals available. UnityFilms gives the same 70% healthy bar for 2026. The reference figure of 73.6% is vidIQ's own Shorts average, which sits just above the healthy bar and gives the comparison on every Shorts result a real channel to measure against.

      Length still matters inside Shorts, and the checker carries two warnings for it. The analytics company 1of10, reporting community-observed working ranges, puts healthy average percentage viewed at roughly 80 to 90% for Shorts under 35 seconds, and 60 to 75% for Shorts over a minute. A 30-second Short at 72% is therefore Strong on the general bar but short of the top range for a Short that short, and the checker says so. A 90-second Short at 65% is Average on the general bar but inside the healthy range for a long Short, and the checker says that too. The middle band, 35 to 60 seconds, has no published range of its own: Vidpal, which reports the 1of10 figures, interpolates 65 to 80% for it and says plainly that the interpolation is its own. This checker does not use the interpolated band, because an interpolated figure is not a published benchmark.

      Shorts have a second retention number that the checker does not rate but that you should read alongside the result: Stayed to watch, the share of viewers who stayed past the opening seconds rather than swiping away. Average percentage viewed is calculated only over the viewers who stayed, so a Short can show a respectable 70% average percentage viewed while most of the people who saw its first frame swiped away. Vidpal's rule for combining the two is to multiply them: 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% Stayed to watch and 75% average percentage viewed scores about 0.53; a Short at 45% and 85% scores about 0.38, even though its average percentage viewed looks better on its own. If your Short rates Strong here but its views have stalled, check Stayed to watch before you conclude the content is the problem: the opening frame and the first spoken line decide whether anyone stays long enough to be counted.

      Shorts can run up to 3 minutes, and the checker enforces that ceiling: a video longer than 3 minutes entered as a Short is rejected with a message pointing to the long-form check instead. The worked comparison that shows why length matters inside Shorts comes from Vidpal: a 30-second Short at 85% holds the average engaged viewer for about 25.5 seconds, while a 60-second Short at 65% holds them for 39 seconds. The longer Short keeps people longer even though its percentage is lower. Seconds are the number to watch when you are choosing a length, because a longer Short that holds a lower percentage can still deliver more watch time per viewer.

      TikTok benchmarks

      TikTok retention is measured as completion rate: the percentage of viewers who watch the video from the first frame to the final frame. TikTok Analytics reports it per video, alongside average watch time in seconds. Completion is the metric the For You Page weighs most heavily, ahead of re-watches, shares, comments, saves, and likes, which is why the checker rates it rather than any engagement figure. The table below is Viralo's 2026 completion-rate table, the only complete tiered table for TikTok published by length band.

      Video lengthBelow averageAverageStrongExceptional
      15 to 30 secondsUnder 40%40% to 54%55% to 69%70% or more
      30 to 60 secondsUnder 30%30% to 44%45% to 59%60% or more
      60 seconds or longerUnder 20%20% to 34%35% to 49%50% or more

      The reference figure of 38% is Socialinsider's 2026 average completion rate across about 70 million posts, and Socialinsider puts top-quartile creators at 55% or more. That 55% figure lines up with Viralo's strong floor for the 15 to 30 second band, which is a useful cross-check: two independent sources put the same number at the boundary between average and strong for short TikToks. Viralo also puts average first-3-second drop-off on TikTok at 45%, which is why the fix sentence on a below-average TikTok result points at the first 2 seconds before anything else.

      Two lengths fall outside Viralo's table, and the checker handles each honestly. Videos under 15 seconds have no published tiered row, so the checker rates them against the 15 to 30 second band and warns that the bands are a floor, not a target: TT Calculator puts typical completion near 89% for 7-second clips and near 73% for 15-second clips, so a 10-second video rated Strong at 60% is still below what short clips typically achieve. Videos of 3 minutes or longer sit inside the 60-second-or-longer band, but completion keeps falling with length inside that band: TT Calculator puts typical completion near 17% at 3 minutes, near 12% at 5 minutes, and near 8% at 10 minutes. A 3-minute TikTok at 25% is therefore Average on the band and well above the typical figure for its exact length, and the checker carries a warning saying so, because the band alone understates the result.

      The length curve is the steepest on TikTok of any platform the checker covers. TT Calculator's 2026 estimates run from about 89% completion for 7-second clips, to 73% at 15 seconds, 55% at 30 seconds, 38% at 60 seconds, 28% at 90 seconds, 17% at 3 minutes, 12% at 5 minutes, and 8% at 10 minutes. Every additional 15 seconds of length costs roughly 12 to 18 percentage points of completion in the sub-60-second range. That is the arithmetic behind the checker's cross-length warning for TikTok, and it is also why longer TikToks are judged on total watch time as well as completion: a 3-minute video at 17% completion still delivers about 30 seconds of average watch time per view, against about 11 seconds for a 15-second video at 73% completion. The short video wins the percentage and the long video wins the minutes, and the algorithm weighs both.

      TikTok videos over 10 minutes are outside this checker's bands, because 10 minutes is the general upload ceiling the benchmark sources assume, and no published tiered table covers longer videos. The checker rejects a length over 10 minutes rather than rating it against a band built for shorter videos. If you post longer TikToks, compare them against your own previous videos of similar length, which is the fairest comparison at any length where published benchmarks run out.

      Why length decides everything

      The core insight of this checker is a single sentence: never compare retention across lengths. The evidence for it is unusually strong. In a study of 5.3 million YouTube videos published at ICWSM 2018, Wu, Rizoiu, and Xie found that duration alone explained over 58% of the differences in the share watched. Length is not one factor among many. It is most of the explanation. A creator who compares a 3-minute video with a 30-minute video on the same percentage is mostly measuring the clock, not the content.

      The arithmetic makes the point concrete. A 35% retention rate on a 30-minute video represents 10.5 minutes of watch time per viewer. A 60% retention rate on a 3-minute video represents 1.8 minutes. The shorter video has the better percentage and dramatically less raw watch time. YouTube's systems use both average view duration and average percentage viewed as ranking signals, so the long video with the lower percentage can still be the stronger performer in the terms the algorithm cares about. The checker prints the average view duration on every result for exactly this reason: the percentage alone is half the story.

      The same logic runs through Shorts and TikTok. 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. On TikTok, a 15-second video at 73% completion delivers about 10.9 seconds of average watch time per view, while a 3-minute video at 17% completion delivers about 30.4 seconds. In each case the shorter video wins the percentage and the longer video wins the minutes. Neither number is the whole truth on its own, and a tool that rated only the percentage would punish every creator who made a longer video.

      The practical rule that follows is short. Compare each video with your own previous videos of similar length first, and with the published bands for that length second. YouTube Studio does the first comparison for you: the Typical retention range on the retention graph is built from your 10 latest videos of similar length, and it is the fairest yardstick available, because it holds your channel, your audience, and your format constant. The published bands in this checker are the second comparison: they tell you where your length band sits on the wider platform, so a video that is typical for your channel can still be recognized as strong for its length, or the other way around.

      Length also decides how much of the first minute matters. Retention Rabbit found that more than 55% of viewers leave within the first minute regardless of video length, and Alan Spicer found that the average YouTube video loses 20 to 30% of its viewers within the first 30 seconds. On a 3-minute video, the first minute is a third of the whole video. On a 30-minute video, it is a thirtieth. The same early loss therefore costs a short video far more of its percentage than a long one, which is one more reason the bands fall as length rises, and one more reason the opening deserves the attention this page gives it.

      Measured medians vs editorial targets

      There are two kinds of retention benchmark in circulation, and they disagree. The first kind is measured: medians and quartiles computed from real analytics data across thousands of videos. The second kind is editorial: healthy ranges assembled by experienced creators and consultants from audits, client work, and platform observation. Both kinds are useful, and they answer different questions. A median tells you what videos actually do. A healthy range tells you what good channels aim for. Confusing the two is the most common way creators misread their own numbers, so this section sets out both, with the disagreement stated plainly.

      The best measured data available is AIR Media-Tech's, published by Monvilo: Studio data from 302 partner channels and 21,755 long videos with 1,000 or more views, published in 2025. The medians are 38% for 3 to 5 minute videos, 30% for 5 to 15 minutes, 26% for 15 to 60 minutes, and 17% for videos over 60 minutes. The top-quarter thresholds are 46%, 41%, 33%, and 25% for the same bands. vidIQ's own numbers point the same way: its channel averages 30.3% average percentage viewed across 16.5 million long-form views, and it calls roughly 30% a realistic good target for a 6 to 9 minute video, noting that third-party studies put the platform average in the low-to-mid 20s.

      Video lengthMedian average percentage viewedTop 25% of videos
      3 to 5 minutes38%46% or more
      5 to 15 minutes30%41% or more
      15 to 60 minutes26%33% or more
      Over 60 minutes17%25% or more

      Set those medians against the editorial ranges this checker rates against, and the gap is clear. For a 4-minute video, the measured median is 38% and the editorial healthy range is 50 to 70%. For an 8-minute video, the measured median is 30% and the editorial strong floor is 40%. Monvilo states the conflict directly: popular bands like 50 to 70% for videos under 5 minutes are blog guidance, not measurements, and sit well above the medians. vidIQ makes the same point from the other side, calling 50% or higher benchmarks inflated and putting a realistic good target for a 6 to 9 minute video at roughly 30%.

      This checker rates against the editorial tables, and it does so for two stated reasons. First, the editorial tables are the only complete tiered tables: they give a floor for every tier in every band, which is what a rating tool needs, while the measured data gives medians and one quartile threshold per band. Second, the editorial tables are what the creator-economy sources the spec is built on actually publish, and the platform average of 23.7% and the under-5-minute strong range of 50 to 70% are the figures creators ask about. But the checker does not hide the medians. Every long-form result carries them in the notes, so a creator rated Average can see that the rating already sits above the measured median for the band, and a creator rated Below average can see how far the number sits from what most videos actually achieve.

      The two kinds of benchmark also agree on more than they disagree about. Spicer's average floors sit at or just above the AIR medians in every band: 40% against a 38% median for under-5-minute videos, 35% against a 30% median for 5 to 10 minutes, 30% against a 30% median for 10 to 20 minutes, and 25% against a 26% median for videos of 20 minutes or longer. An Average rating from this checker therefore means roughly what it says: at or above the middle of what videos actually do, and below the level good channels aim for. The tiers are honest about their nature. They are targets assembled from creator-economy expertise, checked against measured medians, and labeled as such on every result.

      One gap in the measured data is worth naming. There is no reliable public median for Shorts. Monvilo notes that the documentary channel in its article holds 85% on its Shorts and vidIQ reports 73.6% on its own, and calls that the extent of the reliable public evidence. The Shorts thresholds in this checker are therefore editorial throughout, and the checker says so. A Shorts rating is a judgment against creator-reported working ranges, not a percentile against a measured population, and it should be read that way.

      Where to find your numbers

      The checker rates the number you type in, so the number has to be the right one. Each platform reports retention under a slightly different name and in a slightly different place, and the wrong figure gives a wrong rating. This section tells you exactly where each number lives.

      For a YouTube long-form video, open YouTube Studio, click Content in the left menu, select the video, and open the Analytics tab. The Audience retention graph is at the bottom of the video analytics page, and the Engagement tab shows retention across multiple videos. Average percentage viewed is the figure to copy: it is the share of the video watched, and it is the metric the long-form benchmarks were built for. Average view duration is the same fact in minutes and seconds, and the checker works it out from your percentage, so you do not need to enter it separately. Two rules govern the report: retention data takes 1 to 2 days to process, and the video must be at least 60 seconds long with at least 100 views before key moments appear. A video shorter than 60 seconds may show no retention graph at all, and the checker carries a note saying so when the length you enter is under a minute.

      For a YouTube Short, open 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 the content list, where you will find Stayed to watch, average view duration, average percentage viewed, and the retention curve. Copy average percentage viewed into the checker. Read it together with Stayed to watch, because average percentage viewed counts only the viewers who stayed past the opening seconds: a Short can show a healthy percentage while most of the people who saw its first frame swiped away. Wait at least 48 hours after posting before judging a Short, because the first test audience is small and early percentages swing.

      For a TikTok video, open TikTok Analytics and find the video under the Content tab, or open the video's own analytics from your profile. Completion rate is the share of viewers who watched to the end, and average watch time is the mean seconds watched. Copy the completion rate into the checker. If TikTok shows you average watch time in seconds instead, divide it by the video length in seconds and multiply by 100 to get the percentage: 15 seconds of average watch time on a 30-second video is 50%. Note that average watch time as a percentage of length and completion rate are different numbers on TikTok, and the benchmarks in this checker are completion-rate benchmarks. A video can have a high average watch time percentage and a modest completion rate when many viewers watch most of the video without finishing it.

      One more timing note applies to all three platforms. Retention figures move in the first days after publication as the algorithm tests the video with wider audiences, and early viewers are usually your most loyal audience, which flatters the number. Judge a video on its settled figures, a week or more after posting, and compare it with your own previous videos at the same age. A rating made on day-one data is a rating of your subscribers, not of the video.

      How to improve retention

      A rating is only useful if it points at a fix. The fixes below are ordered by leverage, and each one is tied to the part of the retention curve it moves. They are drawn from the same sources as the benchmarks, so the advice and the numbers come from the same evidence.

      Fix the opening first. The first 30 seconds decide most of what happens next. Retention Rabbit found that more than 55% of viewers leave within the first minute regardless of video length. Alan Spicer found that the average YouTube video loses 20 to 30% of its viewers within the first 30 seconds. vidIQ puts 60% still watching at 30 seconds as the target for a good hook. The common mistakes are a branded cold open, a generic greeting, a call to subscribe before any value has been delivered, and an intro that repeats the title without adding anything. Spicer's hook formula has three parts: a pattern interrupt in the first 2 to 3 seconds, a specific value promise, and a curiosity gap that can only be resolved by watching on. For Shorts and TikTok, the window is shorter: the first 3 seconds decide whether the viewer stays or swipes, and Viralo puts average first-3-second drop-off at 45% on TikTok and 55% on Shorts.

      Deliver on the title promise early, then go deeper. Saving the main answer for the end of the video backfires: viewers who feel strung along leave frustrated and are less likely to click the next video. Give a clear, concise answer to the title question within the first 2 to 3 minutes, then spend the rest of the video on advanced techniques, examples, and nuance. The viewers who came for the answer stay for the depth, and the viewers who only wanted the answer leave satisfied, which is a better outcome than a frustrated exit.

      Use pattern interrupts every 60 to 90 seconds. Attention wanders after about a minute of the same stimulus. The most retention-optimized creators build in deliberate changes at regular intervals: a camera angle shift, a graphic, a music change, B-roll replacing a talking head, or a text overlay emphasizing a key point. Variety stops the passive click-away. On TikTok videos over 30 seconds, Viralo's advice is to change the visual pattern every 5 to 8 seconds.

      Stack open loops. An open loop is a reference to information coming later in the video. Each one creates a small commitment to keep watching. The best creators stack them: close one loop while opening another, so there is always an unresolved question carrying the viewer forward. The classic form is a numbered list with the strongest item teased early: the third technique is the one that doubled retention, and it comes at the end.

      Match the length to the topic. If the value fits in five minutes, stop at five minutes. Padding creates dips. If the value genuinely continues, keep going: a 30-minute video holding 35% delivers 10.5 minutes of watch time per viewer, more than five times what a 3-minute video at 60% delivers. YouTube's own documentation says there is no optimal length for YouTube videos. The right length is whatever the content supports without padding, and the checker exists to judge the result at whatever length you chose.

      Study your own spikes and top moments. The retention graph marks top moments, where almost no one left, and spikes, where viewers rewatched or skipped back. Those are the parts of your own work that hold people. Replay them, find what made viewers stay, and repeat it. The same graph marks dips, where viewers skipped ahead or left. Go to the exact timestamp of each dip and watch those 30 seconds: you will almost always find a specific cause, a tangent, a slow stretch, or a transition that lost energy. Cut the cause or add a pattern interrupt before it in the next video.

      Compare against yourself, weekly. Track the average retention across your videos each week, your best video against your worst, and the first-3-second retention on short-form. If the average rises even 5% month over month, the algorithm notices. Consistency of improvement matters more than occasional spikes, and the fairest comparison is always your own trailing average over your last five videos of similar length.

      If the video is ready to earn, the next step after retention is monetization: our YouTube Monetization Checker shows whether a channel meets the Partner Program bars, and our YouTube Money Calculator estimates what a channel at your size can earn. For the upload itself, our Video Length and Upload Spec Checker checks a video against YouTube and TikTok length limits before you post, and our YouTube Title and Description Checker checks the title and description against the character limits that decide what viewers see. For the engagement side of the same video, our Engagement Rate Calculator works out the rate from likes, comments, shares, and views.

      What this checker does not do

      The checker rates one number against published length bands. It does not read your analytics, it does not watch your video, and it does not know your niche, your audience, or your traffic sources. A tutorial that answers a specific question will lose viewers the moment the answer is given, and that is a satisfied viewer leaving, not a failure. A video served mostly to new viewers through Browse will retain differently from the same video served to subscribers. The checker cannot see any of that, and no rating it gives should override what your own retention graph shows about where viewers actually leave.

      The checker also does not predict distribution. Retention is one signal among several: YouTube layers satisfaction signals on top, including post-watch surveys, repeat viewing, session continuation, shares, and comment sentiment. A video with strong retention and weak satisfaction can stall, and a video with modest retention and high satisfaction can travel. Treat the rating as a diagnosis of the number, not a forecast of the video.

      Finally, the checker does not adjust for niche. Educational how-to content averages 42.1% retention and vlogs average 21.5%, on Retention Rabbit's figures, and no published table breaks the tiers down by niche and length together. If your niche sits far from the platform middle, the bands are a rough guide and your own channel history is the better one. The 180-day review cycle on this page exists to catch changes in the benchmarks themselves: platforms change their metrics and their counting rules without notice, and a benchmark from 2025 is a snapshot, not a constant.

      Sources

      All figures on this page were verified on 21 November 2026 against the publisher's own page. Every benchmark is attributed to a creator-economy study or analytics platform. Neither YouTube nor TikTok publishes official retention targets.

      • Retention Rabbit 2025 benchmark report, as published by TubeLab: platform average 23.7%, more than 55% leaving within the first minute, 5 to 10 minute videos best at 31.5%, healthy ranges by length, and niche averages of 42.1% for educational how-to and 21.5% for vlogs. TubeLab: YouTube Audience Retention in 2026.
      • Alan Spicer, YouTube Audience Retention: the audit table of below-average, average, strong, and exceptional bands by length, from hundreds of channel audits and data accessed during his time on vidIQ's Creator Success team. Alan Spicer: YouTube Audience Retention.
      • AIR Media-Tech, Studio data from 302 partner channels and 21,755 long videos published in 2025, as published by Monvilo: medians of 38%, 30%, 26%, and 17% by length band, and top-quarter thresholds of 46%, 41%, 33%, and 25%. Monvilo: What Is Good Audience Retention on YouTube?.
      • Wu, Rizoiu, and Xie, ICWSM 2018: duration alone explained over 58% of the differences in the share watched across 5.3 million YouTube videos, as cited by Monvilo.
      • vidIQ: roughly 30% average percentage viewed as a realistic good target for 6 to 9 minute long-form videos, a 30.3% channel average across 16.5 million long-form views, a 73.6% Shorts average over the 12 months to June 2026, and 60% still watching at 30 seconds as the hook target. vidIQ: YouTube Audience Retention, What Is a Good Rate?.
      • Humble and Brag, YouTube Shorts Benchmarks 2026: healthy average percentage viewed of 70% or above for Shorts, below 50% a structural problem, above 100% replays, and 80% in the first 3 seconds with 60% through the midpoint as the retention shape. Humble and Brag: YouTube Shorts Benchmarks 2026.
      • 1of10 community-observed Shorts ranges, as reported by Vidpal: 80 to 90% healthy for Shorts under 35 seconds, 60 to 75% for Shorts over a minute. Vidpal: Good Retention Rate for YouTube Shorts (2026).
      • Viralo, Video Retention Rate Benchmarks 2026: the TikTok completion-rate table by length band, first-3-second drop-off of 45% on TikTok and 55% on Shorts, and the 15 to 21 second sweet spot for TikTok completion. Viralo: Video Retention Rate Benchmarks 2026.
      • Socialinsider 2026 social media benchmarks report, as published by PostEverywhere: average TikTok completion rate of 38% across about 70 million posts, with top-quartile creators at 55% or more. PostEverywhere: TikTok Metrics and KPIs.
      • TT Calculator, TikTok Video Completion Rate by Video Length, updated July 2026: typical completion from about 89% for 7-second clips to under 10% for 10-minute videos, and the canonical bands of 80% or more excellent, 60 to 80% good, 40 to 60% average, and under 40% weak. TT Calculator: TikTok Video Completion Rate by Video Length.

      Frequently asked questions

      What is a good audience retention rate on YouTube?

      It depends on the video length, and there is no single good number. The platform average is 23.7%, from Retention Rabbit's 2025 benchmark report covering more than 10,000 videos. For videos under 5 minutes, 50 to 70% is the healthy range. For 5 to 10 minutes, 40 to 55% is solid. For 10 to 20 minutes, 35 to 45% is good. For videos over 20 minutes, 25 to 35% is respectable. vidIQ puts a realistic good target for a 6 to 9 minute video at roughly 30% average percentage viewed, and its own channel averages 30.3% across 16.5 million long-form views. The fairest comparison is your own last five videos of similar length, which YouTube Studio shows as the Typical retention range.

      What is a good retention rate for YouTube Shorts?

      A healthy average percentage viewed for Shorts is 70% or above. Below 50% indicates a structural problem: the hook is failing, the pacing drops off midway, or the content does not deliver on what the opening promised. Above 100% means viewers are replaying the Short, which is one of the strongest positive signals available. Length changes the bar inside Shorts: the analytics company 1of10 puts the healthy range at roughly 80 to 90% for Shorts under 35 seconds, and 60 to 75% for Shorts over a minute. For reference, vidIQ's own Shorts averaged 73.6% average percentage viewed over the 12 months to June 2026. Never compare a Short with a long-form video: Shorts loop and last seconds, so their percentages run far higher.

      What is a good completion rate on TikTok?

      It depends on the video length. Viralo's 2026 table puts strong completion at 55 to 70% for 15 to 30 second videos, 45 to 60% for 30 to 60 second videos, and 35 to 50% for videos of 60 seconds or longer. The average TikTok completion rate is 38% across all posts, from Socialinsider's 2026 analysis of about 70 million posts, and top-quartile creators hit 55% or more. Completion falls fast with length: TT Calculator puts typical completion near 89% for 7-second clips, near 73% at 15 seconds, near 55% at 30 seconds, near 38% at 60 seconds, and near 17% at 3 minutes. Judge a TikTok only against videos of a similar length.

      Why is my retention lower on longer videos?

      Because length is the main thing that decides retention. In a study of 5.3 million YouTube videos published at ICWSM 2018, Wu, Rizoiu, and Xie found that duration alone explained over 58% of the differences in the share watched. Every extra minute is another chance for someone to leave, so a longer video will almost always show a lower percentage than a shorter one from the same channel. That is normal, and it is why this checker rates each video against the band for its own length. A 35% retention rate on a 30-minute video is 10.5 minutes of watch time per viewer, while 60% on a 3-minute video is only 1.8 minutes: the longer video holds each viewer far longer despite the lower percentage.

      Is 50% retention good?

      It depends on the length. On a video under 5 minutes, 50% is the bottom of the strong band: good, not great. On a 5 to 10 minute video, 50% is comfortably strong. On a 10 to 20 minute video, 50% is strong and close to exceptional. On a video of 20 minutes or longer, 50% is strong and only 5 points from exceptional. On a Short, 50% is the bottom of the average band: watchable, but below the 70% healthy bar. On a 30-second TikTok, 50% is strong. The same number means five different things at five different lengths, which is why a single good-retention number does not exist.

      What is the difference between average view duration and average percentage viewed?

      Average view duration is the raw time watched, in minutes and seconds. Average percentage viewed is the same fact as a share of the video's length. A 10-minute video with a 4-minute average view duration has 40% average percentage viewed. The two numbers answer different questions: the percentage tells you how much of the video viewers watched, and the duration tells you how much watch time each view delivered. Two videos can share a percentage and deliver very different watch time: 40% of a 30-minute video is 12 minutes per viewer, while 40% of a 5-minute video is 2 minutes. YouTube uses both as ranking signals, so read them together. This checker prints both on every result.

      Where do I find my retention numbers?

      For YouTube, open YouTube Studio, click Content, select the video, and open the Analytics tab. Average percentage viewed sits with the Audience retention graph, and the Engagement tab shows retention across videos. Retention data takes 1 to 2 days to process, and a video needs at least 60 seconds of length and 100 views before key moments appear. For a Short, open Analytics, open the Content tab, and choose the Shorts chip, or open a single Short's analytics for its average percentage viewed and Stayed to watch. For TikTok, open TikTok Analytics and find the video under the Content tab, where completion rate and average watch time are listed per video. Wait at least a week after posting before judging a video: early viewers are usually your most loyal audience, which flatters the number.

      Are these benchmarks official YouTube or TikTok figures?

      No. Neither YouTube nor TikTok publishes official retention targets, and no benchmark on this page should be read as one. Every figure is attributed to the creator-economy study or analytics platform that published it: Retention Rabbit's 2025 benchmark report, Alan Spicer's audit table, AIR Media-Tech's measured medians, vidIQ's channel data, Humble and Brag's Shorts benchmarks, 1of10's community-observed ranges, Viralo's TikTok table, Socialinsider's 2026 analysis, and TT Calculator's length curve. The sources section lists each one with a link. Where the sources disagree, this page says so and names which source governs the rating.