13:08Brian G. Johnson TV · Tutorial
The Fastest Way to Get YouTube to Recommend Your Video
A 13-minute breakdown of why relevance, not quality, decides which videos YouTube keeps pushing.
October 8thThe argument in one line.
YouTube doesn't reward video length directly, it rewards average view duration, so a longer video with mediocre retention can outperform a shorter video with excellent retention.
Read if. Skip if.
READ IF YOU ARE…
- A YouTube creator who keeps hearing retention advice and the 70% rule but can't tell if it actually applies to a channel their size.
- Someone deciding how long to make their next video and wants a number grounded in platform data instead of a guess.
- A creator whose videos get decent retention percentage but still don't get recommended, and wants to know why.
SKIP IF…
- You're optimizing for a platform other than YouTube. This is YouTube-specific algorithm behavior.
- You already track average view duration and watch-time-per-impression as your primary metrics. This covers ground you already know.
The full version, fast.
YouTube's own advice says to keep viewers watching, which pushes creators toward chasing retention percentage, a trap best illustrated by the '70% retention' rule that's unattainable for most channels. The real driver is average view duration, total minutes watched, not percentage of the video completed. A 21-minute video with 27% retention can beat a 6-minute video with 41% retention because its average view duration is higher. Going under 8 minutes leaves almost no room for error. Subscribers and repeat viewers watch longer in the first hours after publishing, which is the early advantage that triggers wider recommendations. The actionable fix: find your best-performing hourly average view duration and roughly double it to set a target length.
Free for members
Chat with this breakdown — free.
Sign in and you get 23 free chat messages on us — ask for the hook, quote a framework, find the exact transcript moment, generate a markdown action plan. Bring your own key when you want unlimited.
Create a free account →Where the time goes.
00:00 – 01:16
01 · Cold open: does length matter
Johnson poses the core question and promises evidence from YouTube itself rather than guesswork.
01:16 – 02:36
02 · What YouTube officially says
YouTube Help and engineer Todd Beaupre both describe the goal as 'keep viewers watching,' a phrase Johnson has seen recur since 2012.
02:36 – 04:27
03 · Retention and the first trap
YouTube's September 2025 blog ties audience retention to reach, but the '70% retention' rule turns that into an unreachable target for most creators.
04:27 – 05:44
04 · What YouTube actually wants
Johnson's own underperforming-by-retention videos still pulled strong views, pointing him toward watch time and ad revenue as YouTube's real incentive.
05:44 – 06:41
05 · Videos have been getting longer since 2005
Average video length on big channels climbed from under 4 minutes in 2006 to over 13 minutes by 2019, though length alone still doesn't guarantee views.
06:41 – 07:51
06 · The one metric that matters: average view duration
Johnson lands on average view duration, total minutes watched, as the single metric that most consistently predicts recommendations.
07:51 – 09:45
07 · Trap one: going too short
A side-by-side of Johnson's own videos shows a 6-minute video with 41% retention losing out to a 21-minute video with only 27% retention, because the longer video banked more total watch minutes.
09:45 – 10:36
08 · How long should your video actually be
Johnson's rule of thumb: look at what's already working for similar channels in your niche, and treat 8 minutes as his own practical floor.
10:36 – 14:02
09 · The unfair advantage: super fans
Subscribers and repeat viewers watch measurably longer in a video's first hours, and that early average view duration functions as a test that decides how far YouTube pushes the video afterward.
14:02 – 14:35
10 · Trap two: no focused audience
Without a specific ideal viewer in mind, a channel can't build the returning audience that drives the early-hour watch-time advantage.
14:35 – 15:58
11 · Finding your ideal length
Johnson's method: take the average view duration from your best-performing early hours and roughly double it to set a target runtime, while noting shorter videos can still work with zero room for error.
Lines worth screenshotting.
- YouTube doesn't reward video length directly, it rewards average view duration, the raw minutes an average viewer spends watching.
- The '70% retention' rule is mathematically out of reach for most channels, and chasing it has ended good creators' momentum.
- A 6-minute video with 41% retention and 95,597 views lost out to a 21-minute video with 27% retention and 21,000 views, because the longer video delivered more total watch minutes.
- Average video length on channels with 100K+ subscribers grew from 3.8 minutes in 2006 to over 13 minutes by 2019.
- The more minutes people watch, the more ad breaks YouTube can run, so total watch time is tied directly to ad revenue.
- Going under 8 minutes removes almost all margin for error: retention and click-through rate both have to be near-perfect to compete.
- Subscribers and returning viewers watch noticeably longer than cold viewers during the first hours after a video publishes.
- A video's first 24 hours of average view duration functions as an early test that determines how widely it gets recommended afterward.
- One video got 2.4 times more views specifically because subscribers watched it longer than usual in its early hours.
- A practical way to set video length: find your best-performing hourly average view duration and roughly double it.
- A niche audience with a consistent viewer identity watches longer over time, even when a single video's retention percentage looks weak.
Average View Duration Beats Retention Percentage
THE REAL METRIC
A video's total average view duration, not its retention percentage or its raw length, is what decides whether YouTube keeps recommending it after the first day.
02What YouTube officially says
- YouTube's own documentation has repeated 'keep viewers watching' as guidance since at least 2012, making it long-standing platform doctrine rather than a new trend.
03Retention and the first trap
- YouTube ties audience retention to reach, but treating retention percentage as a pass-fail target, the way the '70% retention' rule implies, sets an unreachable bar that has stalled real creators.
04What YouTube actually wants
- The incentive behind the algorithm is straightforward: more total watch minutes means more ad placements and more revenue, so the platform optimizes for aggregate watch time over any single percentage metric.
05Videos have been getting longer since 2005
- Average video length on channels with 100K+ subscribers roughly tripled between 2006 and 2019, a trend driven by platform incentives rather than creator preference.
06The one metric that matters: average view duration
- Average view duration, the raw minutes an average viewer spends watching, is the metric most directly tied to how often a video gets recommended.
07Trap one: going too short
- A shorter video needs near-perfect retention and click-through rate to compete, because it has far less room to accumulate total watch minutes than a longer video does.
- Two videos from the same channel can have opposite retention percentages and opposite outcomes: the one with lower retention but a longer runtime delivered more total watch time and won.
08How long should your video actually be
- Benchmark runtime against similar channels already succeeding in your niche before picking an arbitrary target length.
- Treat your own shortest reliable length as a floor, not a goal, since going shorter removes your margin for error.
09The unfair advantage: super fans
- Subscribers and repeat viewers watch measurably longer than cold viewers in a video's first few hours, and that early gap is what triggers wider recommendation.
- A video's performance in its first 24 hours functions as a test: strong early average view duration signals satisfaction and expands who the video reaches next.
10Trap two: no focused audience
- Without a specific recurring viewer in mind, a channel can't build the base of returning watchers that produces the early watch-time advantage.
11Finding your ideal length
- A practical way to set target length: find the average view duration from your best early-performing hours and roughly double it.
- Shorter videos can still work, but anything under about 8 minutes leaves almost no room for error in retention and click-through rate.
Terms worth knowing.
- Average View Duration (AVD)
- The average number of minutes a viewer spends watching a video, used by YouTube as a core signal for recommendations, distinct from retention percentage.
- Audience Retention
- The percentage of a video's total length that an average viewer watches before leaving, shown as a declining curve in YouTube Studio.
- Watch Time
- The total cumulative minutes all viewers have spent watching a video or channel, which YouTube uses partly to decide how many ad breaks it can serve.
- CTR (Click-Through Rate)
- The percentage of people who click a video after seeing its thumbnail and title in a feed or search result.
Things they pointed at.
Lines you could clip.
05:13
“Trying to keep everyone watching can become the nail in the coffin for too many good creators. I've done this many times. Maybe you have too.”
05:37
“The more people watch, the more opportunities YouTube has to place ads and YouTube makes a boatload of money.”
10:23
“It's eight minutes for me. Any time I go shorter than eight minutes, I'm in really a dangerous place. That leaves me zero margin for error.”
13:36
“Subscribers are watching this video longer than usual, contributing to 2.4 times more views.”
Word for word.
Read-along
Don't just watch it. Burn it in.
See every word as it's spoken — crank it to 2× and still catch all of it. The same dual-channel trick behind Amazon's Kindle + Audible.
metaphor
Is YouTube more likely to push longer videos? This is an important question because fact of the matter, imagine, just for a moment, imagine the last video you published 30 seconds longer gets way more recommendations, gets way more love from the algorithm and it was literally the difference between an eight or maybe a 10 minute video and a 10 and a half minute video.
ridiculous and yet at the same time there must be some truth to video link. And I wanted to know, I'm talking evidence, proof, detailed information from YouTube that would guide me so I could really answer the question. And more than that, I could know how to move forward publishing a video that had a far better chance of recommendations.
By the end of this video, I'll break down why longer videos get views, but more importantly, well, exactly how long should your next video be? And more importantly, how to avoid those traps so you can move forward and understand it and really just save time and energy. Hey, if you're new to the channel, my name's Brian.
If you're returning, great to have you. Let's jump in. I started out and I found this statement from YouTube help.
Build anticipation and curiosity. OK, we hear a lot about curiosity, but it's this statement right here. Keep viewers watching.
See, I've been on YouTube for a long time and I keep seeing that type of a phrase showing up again and again. Check this out. This is way back from October 2012.
Keep viewers watching. And then there's Todd Bupre, an employee at YouTube, an engineer working on recommendations. And ultimately, he's one of the principles behind the algorithm.
And he, too, mentioned a bunch of statements that kind of start off with this how long people watch videos. And I started thinking, OK, well, I keep seeing this phrase. how exactly does YouTube measure it because there are just too many detailed information, metrics, it's a mess.
And I can click into any video and find five to eight different metrics that are all supposed to be important and I don't have a clue. Watch time, average view duration, percentage viewed, audience retention, it's absolutely a mess. And right around that time, I found this additional quote from Todd and he says, We look at both the absolute time you spend as well as the relative time.
And I got nothing. Like, what does that mean? I need simple and clear answers.
And then I remembered retention. This is how many viewers are still watching as your video progresses. And it brought me to this.
From the official YouTube blog about a year ago, September 3rd, 2025. High audience retention tells the YouTube system that viewers are finding your content engaging, okay, good, which can positively impact your video's reach. So I'm thinking, okay, I really gotta pay attention to this retention stuff.
But I gotta tell you, this is also where the first trap is sprung. And then there's this famous quote from Mr. Beast from about five years ago.
And I got to tell you, I think Mr. Beast is a genius. But this quote, 70 % retention.
So he's saying you got to keep 70 % of the audience watching to the end of the video. That's not just mission impossible, that's unattainable for the vast majority of YouTube mere mortals, me included. And furthermore, it's typical vanilla YouTube advice that keeps people stuck.
Like focusing on this stuff you keep hearing about. Like for me, I need a new way, a better way, a new way to think about YouTube, and this is why. The typical advice is get people to click and then keep them watching.
Get them to click, get them to stick. And the problem is, is that takes so long. To learn how to communicate on camera, the perfect script, you can spend so much time and energy.
trying to perfect your video, and in the end, it's inevitable people are going to leave. And here's what I came up with. Instead of trying to create the perfect experience, what if I anticipate that people will leave?
And furthermore, check this out. Like some of my best videos, the retention, 41 % retention, 95 ,000 views on a video I published about a month ago. And yet Mr.
Beast is like, well, you need 70%. Goodness. Here's another video.
It's even worse. So many people left and yet the video still performed. Fact is, I don't think this information is helpful.
And sadly, for too many good creators, this is without question . But that brings me to what exactly does YouTube want? Like, if I wanna do well on YouTube, I have to understand what really powers the algorithm and what YouTube is trying to achieve.
And the fact is, the more people watch, the more opportunities YouTube has to place ads and YouTube makes a boatload of money. And the shift has been absolutely unmistakable. And that is the shift in just how long videos are getting.
It might not make sense for you to publish a 20 minute video, but sadly, that's what we have to compete against today. Now back in 2005, the average video length was about five minutes. Five minutes and 20 seconds or so.
Fast forward 2010, the video length is almost approaching six minutes. 5 .7 minutes. 2016, it's over 11.
minutes long and in 2019 holy smokes this is 2019 it's 13 minutes long make no mistake videos are getting longer however it doesn't guarantee that if we publish a longer video we're gonna get the results we're after but on the other hand Todd Bupre once again senior director of growth and discovery 12 years at YouTube Now, he mentioned how long somebody watched was also indicative of their level of satisfaction.
Keep people watching longer. And then this statement from Creator Insider 2025, he says, while we do look at how long people watch videos, again, it's only one factor. that we can consider.
So we have to be aware that there's a number of factors. But the fact is, as I , a longer video is far more likely to win. And here's why.
One single metric that really is responsible and really it's the metric that matters on YouTube today. 2027, and that is average view duration. Now, this comes from YouTube creators official channel, and they mention how long viewers are watching a video.
That's average view duration, which we all get. But videos with higher average view duration have this built in cushion that often triggers the needed watch time metrics. And I want to draw your attention to two of my most recent videos because this clearly spells it out.
This video, six minutes long, 8 ,400 views, three weeks old, currently getting one view a day. It had incredible retention. And yet a video I published four days ago, 20 minutes, 51 seconds, 21 ,000 views.
About 1 ,000 views a day streaming in right now. The video continues to get recommended. Why?
Well, first off, check this out. This is where it gets weird. The average view duration for the winning video, and can I just say I'm not sure what the heck that graphic is?
it's five minutes and 32 seconds. So on average, the audience watched for just over five and a half minutes. It's not that spectacular.
In my six minute video, I didn't account for the fact that people leave. Like, if we're not thinking and if we're not really honoring that so many people leave within literally seconds, it's almost impossible to win today. Now, 29 % of the audience left within 30 seconds.
In other words, the retention was 71%, which is pretty darn good. But still, with 29 % of the audience leaving, . The average view duration was only 3 minutes and 9 seconds.
31 seconds less than usual. And furthermore, this puts a tremendous strain and a lot of pressure on so many other elements like the CTR, satisfaction signals, recommendations, all of it. Now, what about Mr.
Beast? Remember, he says, well, you gotta hit 70%. His videos right now, as I'm filming this video, 57 minutes, 23 minutes, the shortest is 14 minutes long.
Sure seems like videos are getting so much longer. And again, how do we compete? But more importantly, how long should your video be?
Now, the way to answer that question is look at similar channels. And when you do that, you're going to get a clear indication of what's already working for channels that are in your space serving the audience you want to get in front of and more. And I also have to mention that so many channels that we see on YouTube have this massive unfair advantage.
and it absolutely triggers watch time. And I wanna break down how that works so you can duplicate it, so you can gain that unfair advantage for yourself. Now for me, the answer is eight minutes.
Like when we talk about like how long should your video be, few things to consider. Like any time I go shorter than eight minutes, I'm in really a dangerous place.
That leaves me zero margin for error, perfect retention, high CTR. It's gotta be super high on a short video. What if your topic isn't that compelling?
Well, it's almost like it's almost impossible out of the gate unless you go longer. Now, if we go back to the video that was just under 21 minutes, now here's what's so interesting. The CTR in the first 24 hours was average, again, average at best, 5 .6%.
And the retention? Holy smokes, most of the audience bailed in pretty early. But here's the thing that's interesting.
When you really lean in to a specific niche, when you serve an audience, there will always be a percentage of people that over time really begin to appreciate what you do. We'll talk more about how to really leverage that because again, it has this huge impact on the algorithm. And here's how that unfair advantage actually works.
You can see the video's doing fantastic. And a first major test hits around hour four. And the algorithm is really trying to understand how well is the audience receiving this video?
Like how satisfied are they? Now, here's where it's a little weird. The average view duration, in around five minutes and 30 seconds.
But it's really in the first 24 hours that it matters so much. Now the first 24 hours average view duration was five minutes and 44 seconds. But something happened in the first few hours right inside of that four hour window.
You see, in hour one, viewers watched for six minutes and 52 seconds. And then in hour two, over seven and a half minutes. And that set the video up to have exceptional watch time metrics right around that four hour point.
And sure enough, remember, the views absolutely skyrocketed. And the reason this happens is based on what I like to call like super fans or people to a channel that just really appreciate they come back. And it's so interesting because YouTube talks about subscribers or sometimes they mention returning viewers.
They watch longer and substantially longer. So imagine the difference a video has on a channel that really honors that. Well, what happens is, again, this.
They watch for an exceptionally long period of time in the first few hours and the video goes out to a slightly cooler audience but still warm. And basically that's how the algorithm works. The longer the video stays alive, the colder the audience, the metrics come down.
But... what we can do is really, really honor those super fans because they watch longer and they lead to these super views. And check this out, the results, more recommended views.
In my dashboard, this is what it says. Great news, subscribers are watching this video longer, contributing to more than 2 .4 times more views. Now, here's what actually happens.
The algorithm will look at viewers that are engaging when they're subscribers and it'll literally kind of create an avatar based on like the perfect candidate. based on viewers on YouTube that are watching similar content and it'll make sure your video gets connected with them and you're far more likely to grow to bring viewers into your channel and it's just it's a beautiful experience.
However, in order for this to happen, we really want to plan for those views. Like, who are we making videos for?
Now, like imagine if you're into hunting and fishing and someone says, well, I love to fish. And they're thinking Montana fly fishing. And yet you're you're thinking, I want to ice fish.
Those are very different viewers. And furthermore, if we publish and people subscribe for fly fishing and then you give them an ice fishing video, they may not click. And what does that do?
It impacts the video negatively in terms of that watch time per impression, the thing the algorithm rewards, and it's really challenging. But again, how long should you go? Well, remember those screens I shared just a bit ago.
This right here, what I like to do is take a video that's a bit longer, look at the best performing hours, maybe two or three hours, and multiply those by two. So here, seven and a half minutes, that's gonna equate to about a 15 minute runtime. So 15 to 20 minutes is a great place for my videos to be.
Now, at the same time, it's important to value that shorter videos can drive views. My most successful video, only eight minutes. But again, wow, not a lot of room for error.
How exactly can you go longer without padding? Well, go into more detail. Look for specific studies that you can talk about.
Tell a story and really lean into the experience. And over time, storytelling will be something you just naturally get better at. flows.
Now do this. Check out this video on the screen. You dig?
The Hook
The bait, then the rug-pull.
Brian G. Johnson opens with the question every creator eventually asks: does a few extra seconds of runtime change whether YouTube recommends a video. He spent the rest of the video chasing down YouTube's own statements and his channel's real analytics to find an answer that has nothing to do with a fixed length.
Named ideas worth stealing.
15:01model
The 2x Hourly AVD Method
Take the average view duration from your video's best-performing early hours (first few hours after publish) and roughly double it to land on a target runtime for future videos.
Steal forDeciding how long to make a YouTube video for any channel with at least one published video to analyze.
How they asked for the click.
VERBAL ASK
15:55next-video
“Now do this. Check out this video on the screen.”
Soft end-screen gesture toward a suggested next video. No product or channel pitch is spoken in the video itself; the affiliate tool links (Morningfame, TubeBuddy) appear only in the description, not spoken on camera.
MENTIONED ON CAMERA
FROM THE DESCRIPTION
PRIMARY CTAWhere the creator wants you to go next.
OTHER LINKSAlso linked in the description.
Visual structure at a glance.
