I build brands people can't stop thinking about 20 billion views
Joined June 2009
- Tweets2.7K
- Following246
- Followers21.2K
- Likes6K
Absolutely unbelievable but we were scheduled to clean THIS plane for a Clean Girl video
We had gone to Austin's private airport "Asking Millionaires to Clean their Jet For Free"
We didn't know the family but they were the only people cool enough to say yes to a silly video
We were in contact with the manager who was going to let us know next time the plane was in town
RIP, what a tragedy
Called it... 1M views realtime on it right now
There is a popular belief that skill in short form content doesn't translate to long form
This is only the case if you dont understand the principles of psychology
If you're a outlier strategy copy cat that only makes bangers when you rip someone else's ideas then yes of course you can't traverse your skill from one domain to another - you dont actually understand what works, you only have an eye for copying what works! (still a valuable skill but it is not the same)
think of someone like @KevinEspiritu, bro is one of the biggest youtubers, a legit artist, sings, and prob like 6 other things. He understands creative at a level most dont even know exist.
I'm telling you! Study the deepest depths of human nature (evolutionary psychology, biases, impulses) and you WILL be a 0.1% performer that CAN traverse creative domains
My first channel: 1200 videos to get 1M subs
My second channel: 8 videos to get 1M subs
THE HARSH TRUTH ABOUT YOUTUBE,..
Just Make 1000 Videos… And yes 80% of them will be garbage… and 20% will be gold…
Stop trying to have the sexy story or going Viral and blowing up overnight…
Stop trying to “post only bangers bro”.
Your favorite OG YouTubers all had to make 500-1000 long form videos before you even knew who they were.
MrBeast made 460 videos to get to 10K subs and 620 to get to 100K… getting to 1M subs took 760 uploads…
This was also the 3rd time he tried to make it as a YouTuber…
PewDiePie was doing daily content and after 100 uploads had 2500 subscribers, 410 uploads to get to 100K subs 719 Uploads to get to 1M subscribers.
MKBHD uploaded 100 videos to get to 78 subscribers… 500 uploads got him to 100K subs… 625 uploads got him to 1M subscribers…
Markiplier took 820 uploads to get to 100K Subscribers…
And 1361 uploads to get to 1M subscribers…
If at a minimum you want to be in the top 1% of channels… which yes even if you ONLY account for active channels… is 100,000 subscribers and a Silver Play Button… Make 1000 Videos…
Your first 100 Crappy Videos are how you LEARN, not how you earn…
And the 8/20 rule is brutal.
We strive for 1 of 10 videos which is an absolute outlier 90/10 rule…
If you follow that logic… 1000 videos means 100 outliers…
80/20 rule means 200 videos that produced the majority of your views and subscribers.
If you sort any large catalog… if they have close to 1000 video, you’re going to notice the pattern that their most popular 100-200 videos carry the entire channel.
The Creators who get to 1M subscribers with less than 100 videos are unicorns…
Many of them also don’t last…
But even the most legendary YouTubers, the ones that were home grown and started in their bedrooms like a regular peasant…
Not in the offices of Buzzfeed… not coming over from another platform like Vine or TikTok… or a former TV career…
They all had to clear 500-1000 videos to get anywhere…
And that is with arguably a high degree of actual talent they were born with…
At least in term of things like raw creative ability and verbal IQ.
There is a certain amount of humility in the idea of locking in and making 1000 videos…
Just like if I told you the body you want is on the other side of 1M reps of the same 5 body weight exercises… (which likely is true).
All 1000 videos can’t be trash.
You MUST be able to significantly improve over time or dramatically improve through pattern recognition.
That is what the legendary YouTubers all learned.
The solution is to keep giving
In business givers eventually meet other givers and they 1 + 1 = 3
while takers 1 * 0 = 0
One thing I’ve learned as a builder:
I probably won’t casually grab coffee and share my entire marketing playbook for free anymore.
I met another Vietnamese builder for coffee and shared a lot of what I’ve learned about marketing apps, the strategies, distribution, TikTok formats, what actually worked for me.
A while later, he basically copied the app.
Then copied the TikTok format too lol.
If it were some random builder on the internet, I honestly wouldn’t care. Copying happens. That’s part of building in public.
But when it’s someone you sat down with, helped, and shared your experience with personally…
yeah :))
1
9
2,307
You have to make shittier videos
Every film school on earth is teaching a skill that explains less than 1% of whether anyone watches
I have the best data on this that exists and I paid for it the stupid way. I used to want the most expensive production possible. I literally wanted to make more money so I could spend more per video. I spent $150,000 on one video
Then I measured what keeps people watching, moment to moment, across 221 videos and 20 billion views. Camera, lighting, framing, shot type, how many things are in the background. All of it together came to under 1%
Take the same idea. Shoot it once on an iPhone and once for $40,000. They perform the same. The only difference is your ego feels the $40,000 one was better
A video with buttery motion and perfect light LOOKS like quality. You're judging a joke by the quality of the microphone
The entire production industry is built on that 1%. Cinema cameras, colour grading, gimbals, crews. So is every AI video company. Billions of dollars racing to make the part that doesn't matter free
Production has exactly one job: the viewer can see and hear the idea. Past that bar nothing you spend comes back
We shoot everything on an iPhone. A drain getting cleaned, $800, 52 million views in 8 days. The $150,000 video I don't talk about
If you do this for a living, this post ruined your week. Go check your own numbers before you reply
15
1,678
Every dollar you spend on Meta ads is a confession that you couldn't come up with an idea
That's going to upset a lot of people who buy ads for a living, so here are the numbers
A million hours of human attention is about one lifetime, a person watching from birth to 114 without stopping. Paid ads rent that lifetime for roughly $4.5 million. We earn it for about $7,900, with the same people and the same budget we had when it cost us $200,000. Our best video did it for $400
Same attention. Same humans. 500x price difference
An ad is attention you steal. You're the thing in front of what they actually wanted, so they meet you with their arms crossed, and the platform charges you for every arm you uncross. Next quarter it costs more, because they've seen your trick
An idea worth watching is attention they hand you. Nobody charges you for it, and they come back
I post 4 times a month. That gets us more sales than companies spending $100M a year on Meta. We launched a toy by telling our audience it existed, zero ads, 100,000 units in the first month. Buying that much attention would have cost about $262 million at 2025 ad prices
The companies spending $100M a year are guessing with more money than me
Ads have one legitimate use: you have an idea that's already working and you want more people to see it faster. That's a multiplier on something real
Everything else is paying a platform to force people to look at something they'd have scrolled past for free, and the bill goes up every year because forcing gets harder the more you do it
An ad budget is rent. It goes up every year and you own nothing at the end
2
6
1,301
There are two ways to get a person's attention, and one of them makes them slowly hate you
I've spent 7 years measuring this across 20 billion views. It holds up at a party, on a sales call and in a marriage
Way 1: you take it. You're the ad in front of the video they wanted. The billboard in front of the view. The guy at the bar who opens with what he does for a living. They know you forced your way in, so they meet you with their arms crossed, and next time you have to push harder to get the same reaction. It gets more expensive every single time
Way 2: they give it to you. You're the thing on the other side. They came looking, there's nothing to push through, so all of your effort goes into being worth the trip. And when someone takes a great trip they come back and they bring a friend. Every time costs you less and pays you more
The difference is one word: friction
Trust is friction running in reverse
I learned this by getting it wrong for $3.25 million. I was chasing the sale. Pushing harder, pitching more, discounting. I was the friction. You can't force a sale any more than you can force a girl to like you. Chase and she runs faster (and probably calls the cops lol)
Here's what it looks like once the friction is gone. We made a toy. To sell it we told our audience it existed. No launch, no discount, no countdown. 100,000 units in the first month. There was nothing left to push
How to do this with anyone:
1. Give the best thing you have first, with no ask. Everyone is too scared to give away the real stuff, which is why it works
2. Never pitch before they'd have paid for what you just gave them. Greed is visible from space
3. When you do ask, make it the thing they came to you for. The deal that pays well but doesn't fit burns years of trust in a day
Do this long enough and the sale stops being something you do to people. You point at the thing and they say yes
1
7
1,083
If someone had handed me this 7 years ago I'd have skipped about $6 million of mistakes
Genuinely should not be free lol
xitter.twitterapi.xyz/Maxwell_Maher/status/2…
We got the cost of stealing 114 years of people's attention down to $7,900
That was our production cost per million watch hours
A million hours is roughly 114 years of human life, spread across everyone watching
Less than a penny per hour
Here's how to find your number:
Count the full production cost, including staff, editing, props and overhead
Divide it by the hours people watched
Do this per video. Compare videos at the same age so an old upload doesn't get an unfair advantage
Then put the most efficient videos next to the least efficient ones
Which ideas made people stay?
Which expensive choices added nothing?
Where did a simple scene outperform something that took days to produce?
That's where I'd look before hiring another editor or doubling the posting schedule
You can spend less of your team's life earning more of the audience's
2
959
Some people are exceptionally talented at getting hired for jobs they can't do
"I'm passionate"
"I'm a fast learner"
"I always go above and beyond"
An interview gives them plenty of room to describe abilities they haven't demonstrated
One question I like asking is:
"What's the most recent thing you learned?"
Some people give you almost nothing. Others start explaining something they got interested in, figured out, and applied
The best writers I've worked with are interested in writing when nobody is assigning them a script. They're reading, noticing things, trying things
I'd follow the first question with:
What made you look into it?
What did you try?
What went wrong?
Can you show me something you changed because of it?
Now we have something concrete to talk about
A draft they reworked. A small project. A decision they made differently after learning something
It doesn't have to be impressive to an outsider. I want to see how they moved from not knowing to knowing
Then give them a small paid task that resembles the actual job
Use the same brief and time limit for each candidate. After the first attempt, give one specific correction and see what changes
Now "fast learner" is something you can watch
4
987
You can spend a day helping someone and still get called an asshole
Cleaning a stranger's grave is a pretty good example
One person sees someone caring for a grave that has been neglected
Another wants to know who gave her permission to touch it
With Clean Girl, those opposite reactions are part of what makes the idea interesting
The viewer who loves it wants to see the grave cared for. The viewer who objects is deciding whether she crossed a line
Both have a reason to keep watching
Then the work gives us a way to explore the questions underneath it
Who is buried there?
What was their life like?
How did the grave get into that condition?
You can reveal those answers while the cleaning happens, so there's more to follow than a surface getting cleaner
This is why "someone might hate it" is such an incomplete criticism of an idea
What do they hate about it?
A cheerful woman spending her day cleaning a grave for a person she never met is something people can have a real disagreement about
Make the assignment "film something nobody could possibly object to" and you can remove that entire reason to care
I want to understand why someone hates the idea before deciding that their reaction means the idea failed
7
919
We borrowed decades of McDonald's advertising to make people watch a stranger clean a bathroom
They had already made the setting familiar
One of our first Clean Girl concepts was cleaning a McDonald's for free
People knew the place. They knew who normally works there. They knew how strange it would be to walk in and start doing the staff's job
"Why is she doing this?"
"Are they even going to let her?"
The audience could get to those questions without an introduction
That's useful when you're trying to make something interesting to people who don't care about your topic
Start with a place, object or situation they already understand
Then change one expectation
Who's doing the job?
Who gets paid?
Who gave them permission?
Keep enough familiar that a stranger can spot what's strange
The viewer should spend their attention wondering what happens next, not figuring out what you're talking about
1
6
961
57 of our videos have passed 100 million views
The first 3 seconds of every one of them does at least one of these 6 things:
1. Opens on a visual you can't explain yet. A splash in green water, no words. The viewer works out "she's fixing a pothole" on their own, and that little bit of work is what keeps them there
2. Puts a person somewhere they shouldn't be. Sitting in a live road. Walking into a cemetery with a scrub brush. The brain flags "is she going to get in trouble" before it flags anything else
3. Uses a simple shape. We scouted for a round pothole on purpose. A circle reads in a tenth of a second, a jagged mess takes half a second, and half a second is the whole game
4. Fills the frame with one color. Bright pink on grey concrete. A screen that's mostly one color holds better than a busy one. It's one of the most reliable things we've found
5. Makes a promise it can pay. The first shot tells you what you're about to see. The biggest retention cliff we've measured is the moment a video looks like it won't keep that promise, usually around second 8
6. Hands you a question instead of a topic. "He spent $100,000 to become a CEO" is a topic. It's the first line of the most expensive video I ever made and it did 10,000 views. A girl kneeling at a grave nobody has visited in years is a question. “Whose is it? Why her?”