15:17Dave Jeltema · Essay
The Six Rules Behind How YouTube Decides Who Sees Your Videos
A business-owner's guide to treating YouTube as a matchmaker with rules instead of a lottery.
October 26thThe argument in one line.
As AI makes polished videos cheap and platform rules keep moving, the only durable edge on YouTube is being verifiably human and trusted by an audience you own.
Read if. Skip if.
READ IF YOU ARE…
- A business owner running a YouTube channel who wants it to bring in the right customers a year from now, not just views.
- A solo creator trying to decide which of the new AI editing and research tools are worth adopting and which to ignore.
- A creator worried about the 8,000 watch hour partner program change and wondering whether monetization is still worth chasing.
SKIP IF…
- You want a step-by-step tutorial on a specific tool; this is a strategic overview with short action lists.
- You already run lives, a community and an email list and want advanced tactics for them.
The full version, fast.
Four shifts are changing YouTube: video production is being automated by AI, platform rules and monetization keep moving, viewing is shifting to the TV and Netflix-style libraries, and a flood of AI content and fakes is eroding trust. Tools will stop being an edge because everyone gets them. The response is to decide what stays yours, run every tool through a need test, spend saved time on ideas, build on a lead magnet, email list and offer, deliver what the title promises, watch YouTube on a TV, build series, then go live, build a community and meet people in person. Being verifiably human is the moat.
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 – 00:56
01 · Introduction
Why tools stop being an edge and what the four shifts are.
00:56 – 06:54
02 · Shift 1: Making videos stops being the hard part
Automated graphics and rough cuts, the Dokapon calculator story, A/B testing worries, and the solve-everything trap.
05:11 – 06:54
03 · What has to stay yours
Non-negotiables, protecting fun, the tool filter, and spending saved time on ideas.
06:54 – 09:08
04 · Shift 2: The rules move under your feet
First-frame view counting, stricter moderation, the 8,000 hour threshold, swappable sponsor segments.
09:08 – 11:11
05 · Shift 3: YouTube moves to the living room
TV viewing share, shows and series, Hollywood crossover, and why to watch on a TV.
11:11 – 14:59
06 · Shift 4: Nothing can be trusted
Mass-produced channels, deepfakes, cheap digital products, and the trust problem.
12:35 – 14:59
07 · How to be verifiably human
Go live, build a place for your people, and meet them in person.
14:59 – 17:31
08 · Trust is the moat and the roadmap
What you bring to the table, a recap list, and an invitation to work one-on-one.
Lines worth screenshotting.
- When every creator can produce a polished video, polish stops being the thing that sets you apart.
- Write down the parts of video making you love and the parts you dread, then let tools take only the dreaded parts so the result still sounds like you.
- Run every tool through one question: what piece of the job does it do, and do I need that piece yet?
- A clipping tool is useless without a library to clip from, and DM automation only pays off once hundreds of people ask for the lead magnet.
- Spend the time AI saves you on ideas, and review what is outperforming in your niche every week.
- Counting a view from the first frame makes public view counts easier to inflate, so judge videos by engaged views instead.
- Doubling the watch hour requirement to 8,000 does not put the goal twice as far away, because growth is an exponential curve.
- You are on rented land, so build on a lead magnet, an email list and an offer that no rule change can touch.
- TV is now 20 to 40 percent of how YouTube videos get watched, so check your thumbnails the way a TV shows them.
- One video gets you found, but a library of themed videos is what makes viewers stay and builds trust.
- When anything can be faked, buyers research more and only buy from people they trust.
- The answer to AI fakes is to be verifiably human: go live, build a place for your people, and meet them in person.
- Solving every problem you can solve is a trap; the tools you build for the wrong problems feel like productivity but leave you in the same spot.
Own the human part, let tools do the rest
WHAT TO LEARN
As tools level production and platform rules keep shifting, the lasting advantage is trust built through ownership, a library of work, and visibly human presence.
- Before adopting any tool, write down which parts of making videos you enjoy and dread, so the work that stays yours is a decision rather than an accident.
- Test each tool by asking what piece of the job it does and whether you need that piece yet; skip clippers and DM automation until you have the library or demand.
- Put the hours that automation saves into ideation, reviewing outperforming videos in your niche each week and logging why they worked.
- Rules and monetization thresholds change constantly, so build a lead magnet, email list and offer that sit outside the platform.
- Deliver exactly what the title and thumbnail promise, and judge success by engaged views rather than an inflated public count.
- Check your channel on a TV and group videos into series and playlists so a single find turns into a binge.
- With AI fakes and clone channels rising, buyers lean on trust, so show proof of real expertise and character.
- Go live in one repeatable format, give your community a warm onboarding, and meet individual people in person to stay verifiably human.
Terms worth knowing.
- Engaged views
- Views where the viewer actually watched and responded, as opposed to the public view count. Jeltema says the algorithm still optimizes for viewer satisfaction, not the public number.
- Hype feature
- A YouTube feature that lets viewers give a video a boost, with three free hypes per week. Videos YouTube sees as misleading lose access to it.
- Lead magnet
- A free resource offered in exchange for an email address, used to build a list you own outside the platform.
- Faceless channel
- A channel that publishes AI-generated or stock-footage videos without a real on-camera person, usually run as a high-volume numbers game.
Things they pointed at.
08:00toolManyChat
14:40toolDiscord
Lines you could clip.
04:22
“The barrier didn't just get lower. It's entirely gone.”
08:50
“You're on rented land, so build on the part nobody can change on you.”
16:30
“When anything can be faked, people will second guess everything.”
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.
If you want YouTube to bring you the right people a year from now, you need to see what's coming, because the game is changing around you. Most people are watching the tools, the new AI features, the new buttons in YouTube Studio, but those are about to stop being an edge because everyone's getting them. So I'll show you the four biggest shifts coming to YouTube.
That's already here, what's on the way, and where it all ends. So you know exactly what to do now before it gets messy. By the end, you'll leave with the one thing none of it can take away from you.
So I've been on YouTube for over 15 years. I've watched it go through every era. First, it was about personality, then subscribers, then search and SEO, and then TikTok showed up and everything became about the almighty algorithm.
And now AI is here. And nobody can predict exactly where all of this goes. But a lot of what's happening has a natural conclusion.
My own channel was built for the way things used to work. It still works, but I'm seeing the writing on the wall. So I'm putting all this into practice myself.
The first big shift is the work itself. If you make videos, you know how long they can take. Scripting and filming, then cutting out every flub and every dead second, then captions and graphics, and by the time you're halfway done, you never want to look at it again.
That last part is going away. Here's a fun little story that kind of illustrates my point. Not too long ago, I was playing an older game on Steam with a friend.
It's called Dokopan Sword of Fury. It's an obscure Japanese -only sequel. to one of my all -time favorites or it was it recently got an english translation so there's barely any documentation of it online and without nerding out about it too much And without nerding out about it too much, the point is, at one point, I hit an enemy for a thousand damage when everyone has like a hundred HP.
It seemed like a buck. That's just way too much damage. Then I used the same move again in the next combat on a different enemy and it did one damage.
I thought the combat would work like the game I knew. I was very confused. And so I started digging.
By the end of that play session, I had built my own calculator that reverse engineered the combat system. I got to turn my stats and the enemy stats, and it would tell me how much damage I would do. It took me less than a few hours, and I did it between my turns.
The barrier didn't just get lower. It's entirely gone. Video is now the same.
When my script is done, I hit a button and say go, and it builds all of my graphics using my own design system. Every list, every pull quote, every title card, all those video assets get built, and then they just sit there waiting to be added to a timeline. Then once I have the rough cut, I say go again.
It takes every one of those assets and places it where it belongs on the timeline. So it's not one button yet, but in the next couple of videos, I'll be having it doing my rough cut too. So one button, boom, video 80 % done.
There are systems and other tools as well that help you write your video in a few hours. Teleprompter apps, subscriptions for music and sound effects, outlier research tools like Velio. I can tell an AI agent to go research anyone in my space.
And two hours later, I have their audience, their products and everything they've ever tried. I watched a short the other day where a guy explained how his company took 30 or 40 ,000 of their editors' copy edits and turned them into a skill. And now anyone on their team can ask for a pass in her style before she ever sees the piece.
It saves her a lot of time when she's reviewing their work. That's just the tip of the iceberg of what's already here. And YouTube is building more of it right into their studio.
On YouTube, you'll be able to ask for feedback on your script and hook. It'll write your titles for you. Thumbnails will be matched to...
each viewing experience automatically. And next year, A -B testing on up to three different cuts of the same video is just going to be an absolute can of worms. I'm not really sure that they thought that one through.
So if you watch a video once and come back later, will it be different? If you send it to a friend, do they see the same one you saw? Will you see A -B tests where one video is five minutes and the other is 50?
What about chapters? Do you need to make a different description for each one that you test? What if a comment was about a specific moment that never happened in the winning video?
And then picture someone with a team. It's not that they just make a different hook. They can film two completely different videos and let you two pick the winner.
So you'll see a new level of competition at the high end. But you don't need a team to be in the game. I'm a one -man band right now, and the tools that let you compete already exist.
There is a trap in this, though. If you have a brain like mine, every problem suddenly looks solvable, so you try to solve all of them. just like that combat calculator for Dokopan.
I have made so many tools and projects that haven't seen the light of day. Solved the problems, but they weren't the right problems that needed to be solved. It feels like productivity, and then you look up months later and wonder why you're still in the same spot.
Keeping your time on the right problems is what I help business owners with one -on -one. If you want to work together, there is a link to apply in the description. And when everyone can make a polished video, polish stops being the thing that sets you apart.
So here's what you can do about it. First, ask yourself some hard questions. Before you hand anything to a tool, figure out what has to stay yours.
Which part of the video making process do you actually enjoy? And which part do you dread the most? If a tool did those parts for you, would the video still sound like you?
Would it be yours? Your answers are your non -negotiables. Write them down.
Those parts stay yours, always, no matter what the tools can do. Then, protect your fun. Don't save it for the end as a reward, build it into the process.
The topics that excite you, the formats you enjoy, the experiments you want to try, that's what keeps you showing up. Rule zero is follow the fun. If it isn't sustainable, then it's not going to last.
Everything else is what the tools are for. I'm not trying to replace myself, but I do want AI to do all the boring parts, so I get to start in a playground. Second, run every tool through one question.
What piece of the job does this do, and do I need that piece yet? You need to set aside the cool things that don't do anything for you yet. A tool that clips your videos is great, but if you don't have a library to clip from, it can wait.
ManyChat is amazing for sending lead magnets through DMs, but it only starts making sense when you have hundreds of people asking for them. And third, spend the time the tools save you on ideas. If there's one thing to actually spend your time on, it's ideas.
They're the lifeblood of success on this platform, and in general. Every week, look at what's outperforming in your niche. Using something like Velio will show you which videos did far better than normal for a channel.
For each one, write down what you think, why it worked. Keep every idea in one place. That way you never sit down to a blank page.
You're always looking forward to the next thing you get to make. The second big shift is the rules. They change every week.
A view now counts from the very first frame, like most other platforms. And whether YouTube meant to do it or not, that encourages clickbait. People like it when numbers go up and turn green.
You don't need to engage the viewer as much to get bigger numbers. Creators will do it subconsciously. At the same time, other rules are getting even stricter.
AI experts on topics like health and money are getting demonetized. And videos YouTube sees as misleading lose their access to the hype feature. Oh no.
Does anyone actually use that? You get three free hypes a week, so you should use yours on this video. Then there's moderation.
YouTube makes an absurd amount of money. But even they can't afford to pay enough people to watch everything that gets uploaded. So it's bots.
And bots throw false positives. There's always more content, and so the bots will always lag behind, and the rules will keep changing. Rules on money are moving too.
Starting February 2027, new channels need 8 ,000 watch hours to join the partner program. That's double what it used to be. At least channels already in it get to keep their spot.
And creators hear that and go, oh no, they moved the goalposts. I'm never going to make it. But growth is an exponential curve.
Doubling the requirement does not put it twice as far away. And if you're a business, ad money was never the point anyway. And brand money is changing shape.
YouTube is already testing sponsor segments that you can swap out. A video sponsored by one brand today can carry a different brand tomorrow. Your whole back catalog can be sold again.
And Studio will soon help you pitch to brands. The money is there, but the rules underneath keep moving. You're on rented land, so build on the part nobody can change on you.
There's three things to do. First, build on what you own. One lead magnet, one email list, one offer.
That's all you need. It gives you a way to reach your people that a rule change can't touch. Second, deliver exactly what your title and thumbnail promise.
This is just human psychology. It's not going anywhere. You'll always need to try and be the better pick over your competition.
Third, judge your videos by engaged views. That's the number YouTube actually cares about. The public count is the one that just got easier to inflate.
But the underlying mechanic of how the algorithm works is still based on viewer satisfaction. The third shift is where people watch. YouTube used to be 50 -50 mobile and computer.
Now TV is 20 -40 % of how your videos get watched. YouTube is now the biggest share of TV viewing in America, at 13 .4%. Short form isn't going anywhere, but YouTube's leaning in.
Shorts are getting grouped into shows, with their own thumbnails and button to watch the playlists like your favorite TV show. You'll be able to reply to comments from your couch by speaking to your remote. Viewers will have different custom home pages with different algorithms by topic.
Put all that together, and you can see that YouTube wants people to watch like they watch Netflix. If somebody finds something they like, they want a library, a chunk of your stuff on one theme, like episodes. And that's how trust gets built.
One video gets you found, and a library is what makes them stay. The ceiling is rising, too. Mr.
Beast took his show to Amazon. It was reportedly a $100 million deal and 50 million people watched in the first 25 days. Markiplier paid for his own movie and it opened to over $18 million at the box office.
YouTubers are heading to Hollywood. But the reverse is true too. Other industries are moving onto YouTube.
News, for example. For the first time, more Americans get their news from social media and video than from TV. 39 % of 18 to 29 -year -olds regularly get news on TikTok.
there will be an even bigger blending across the different platforms. If you think about it, there are some serious A players with real resources jumping on the platform, and they're just getting started. If you don't raise your head, take a look around, your blinders can lead you in the wrong direction.
So here's how to look up. First, go watch YouTube on a TV. Literally, go open it on a TV.
Go find your channel. Look at your thumbnails the way a TV shows them, at a glance. So many people want to grow on this platform, but don't actually watch YouTube.
So you can hear me say this, but until you go see it, it's not going to sink in. Second, give people something to binge. Group your videos into series and playlists by theme.
So when someone finds one of your videos and likes it, a whole show is waiting for them. The fourth shift is where you take all of the things we've already covered and ask, where does it take us? Because even now, if you give me a weekend, I can set up a channel that uploads a YouTube video every two minutes for the next 10 years.
The idea, the title, the thumbnail, an AI voice, the visuals, the edit, the upload, all of it. That's where faceless channels come from. They're playing a numbers game.
Accounts get banned, they find the next trend, and then they do it again. What a fun treadmill. Now imagine, if you don't even have to set it up, you just ask an agent to do it, and voila.
We will see a 10x in the content uploaded to the platform, which will make YouTube raise their prices. But it gets worse. Someone can take your video, reword your transcript, and have an avatar read it on screen.
It's not going to be long before someone deepfakes a popular creator and duplicates their channel. If people know who you are, you're going to have some haters. And they are getting some pretty gnarly tools to be shitheads with.
Digital products are the same. I could throw together a checklist in an afternoon and sell it for 20 bucks. Anyone can, so they're popping up everywhere.
When anything can be faked, people will second guess everything. I don't think it would take me very long to go find a comment section of an indie artist's video where they have to defend themselves against accusations of AI. And when people spend their hard -earned money, they're pickier than ever.
I know I am. I do more research and spend more time on every single purchase. The people I buy from are people I trust.
They've shown me proof I understand their character. So if we put all this together, it sounds pretty bleak. How the heck do you combat against all that?
The answer is to be verifiably human. No one can take that from you. Here's specifically how to do it.
First, go live. How do you do something no one else can? You do it live.
If you have expertise, you can demonstrate it. If you have access to something, you can show it when others can't. It's not that hard.
Pick one format you can repeat. A live Q &A, a walkthrough of how you solve a real problem, a coaching call, or a teaching session on one subject, or try each one. Set the date, tell people about it, and put it somewhere people can sign up.
Send a few reminders as it gets close, rehearse the day before, have a person in the chat helping you out, and if you're going to pitch something, Say so up front. It takes all of the pressure off.
I moved my next live stream up by a month, so you can join me December 10th, and you'll see how I do it. Click the link in the description. I'll send you some more goodies and remind you when the time gets close.
You don't have to change everything you do. I didn't suddenly stop making scripted videos like this one, but I am adding a live layer because it's more human, and I think it's going to matter more and more every year. Second, build a place for your people.
Everyone building a personal brand is building their own little community. People who know who you are and like hearing what you have to say. That place can be your email list.
It can be a free community on school or paid membership on Circle. I use Discord. The platform isn't the important part.
It just needs to be a verifiably human place for people who really want to connect. And don't overlook onboarding. Nobody likes joining something, looking around and asking, now what do I do?
I bring this up because it's something that I need to get better at. So give them a warm welcome. an email sequence, resources to browse course material, or a simple step -by -step guide.
Third, meet your people in person. When people hear that, they think of in -person events, which are great for bigger audiences, but you can meet just one person. And I'm not saying go meet some stranger who has a parasocial relationship with you in a restaurant in another city.
But I've been meeting with people and helping them online weekly for months now. I would love an excuse to meet some of them in person, to have a fun conversation, to take a picture, and honestly, Turn it into some content.
Most people don't do this. And even if they do, they can't do it your way because they aren't you. You don't need to be famous for any of this.
You don't need to dominate the platform. You just need the people you're trying to reach to know who you are. Be human.
All four of these shifts point the same direction. Getting found still matters. Nobody can trust you until they know you exist.
But once they find you, there's a new question. Why should they stick with you over anyone else? It pains me right now to see every business owner starting a channel and following the same old playbook.
They all look the same. And as tools get better, everyone's production rises. Everyone's speed rises.
All the old ways of getting an edge are shrinking. The one edge that isn't affected is what you bring to the table. What you're known for, who you help, the way you talk to them, the stories and the experiences you have to share.
For a long time, I was convinced I couldn't make a channel about growing on YouTube. Other people were already sharing the information. And I was getting my information from them, and they were getting it from somebody else.
It kind of reminds you of a certain shape. I had to go build the experience first, so I'd have something worth sharing. And that's all that I do here, is to help you avoid the same mistakes that I've made.
Anybody can ask ChatGPT how to grow a channel and read the answer to camera. What makes your videos worth watching is your perspective, in service of getting your people what they want. It's not what you say, it's that it came from you.
than a big one that doesn't. That's the moat. It's the one thing none of these changes can take away from you.
Trust. So let's put it all together. Answer the hard questions and write down your non -negotiables.
Check every tool. What piece of the job does it do and do you need it yet? Put the time the tools save you into ideation.
Make this a weekly habit. Build on what you own. A lead magnet, an email list, an offer.
Deliver exactly what your title and thumbnail promise and judge your videos by engaged views. Go watch YouTube on a TV. Think about how your videos work as a series.
Build a place for your people and give them a real welcome. Go live and meet your people in person. You're not too late.
It's not like there's some date where you missed your chance once the clock ticks over. But if you don't act on this, you'll just be waiting for the next video like this to remind you. Most of what gets in the way of succeeding on YouTube is mental and time.
That's the part that I help with. I'm running these experiments on my own channel right now. and I'm taking on 10 business owners to work with one -on -one.
I help you understand the fundamentals that never change. I give you the tools and the direction, and then I'm there to pull you out of your own head when you need it. If that sounds like you and you want YouTube to bring you the right people, the link to apply is in the description.
Or you can just click or tap on the screen to watch this next to see exactly how we would work together and whether it's right for you.
The Hook
The bait, then the rug-pull.
Everyone is watching the new AI features and Studio buttons, but those are about to stop being an edge. Dave Jeltema lays out the four bigger shifts underneath them and the one thing none of it can take away.
How they asked for the click.
VERBAL ASK
17:08product
“I'm taking on 10 business owners to work with one-on-one... the link to apply is in the description.”
Soft pitch placed after the roadmap recap, plus earlier mentions at the solve-everything trap and the live event.
MENTIONED ON CAMERA
Visual structure at a glance.
