My 7,000+ tweets in one knowledge base

You keep treating a prediction engine like a lottery ticket. You upload, you hope, and when nothing happens you blame the algorithm. That's why you can't predict your own results — you never learned what the machine is actually trying to do.

This is the difference between the people who grow and the people who've been "trying YouTube" for three years. It's not talent and it's not luck. One group is running experiments against a system they understand, and the other is buying tickets and getting upset when the numbers don't come up. It is not random, and the moment you understand what it optimises for, the whole platform becomes readable.

1. Watch 10–30 videos in your niche, properly, not skimmed.
2. Let them play through — watch depth is one of the strongest signals available, and finishing tells the system far more than clicking.
3. Verify your phone number and email. Unverified behaviour reads as a consumer, verified reads as a builder.
4. Keep it focused. If the watch history is random the model stays fuzzy. If it's consistent the system gets confident.

1. You stop appearing on the homepage.
2. You get excluded from autoplay trails.
3. New users never discover your niche in the first place.

1. Average retention in your niche.
2. The video length YouTube seems to favour there.
3. The upload frequency the top channels stick to.
4. Title and thumbnail patterns that are driving CTR right now.
5. The emotional tone that consistently gets rewarded.

A personal brand is a single point of failure. If the entire thing depends on one person, then an injury, an emergency or any sudden life change takes the brand down with it. You have built a job that cannot be delegated, paused, or sold.

And it never stops being one. The bigger it gets the more it depends on you personally, so success makes the trap tighter rather than looser. You can't sell it, you can't hand it over, and you can't stop — which means what you've built isn't really an asset, it's a very well-paid job with no notice period.

1. Upload ten videos in your niche.
2. Review the data honestly and find what actually underperformed.
3. Fix that specific thing.
4. Double down hard on whatever outperformed.
5. Upload ten more and run it again, indefinitely.

Stop trying to do everything at once. You will very likely never find success — and can go broke on the way — if you believe your first ten channels should be diversified. You're not spreading risk, you're guaranteeing you're mediocre everywhere.

I understand the instinct completely, because every niche you look at has someone winning in it and it feels like you're leaving money on the table by picking one. You're not. Ten half-built channels in ten niches is ten sets of packaging instincts you never developed, ten audiences you never learned, and ten formats you never nailed down.

1. Build a watchlist of high-ceiling channels — big audiences, trending niches — that are not in your niche but run concepts you could replicate.
2. Check whether the concept has already travelled into other niches. The more places it has been copied, the larger your sample size and the safer the assumption that the concept itself is what works.
3. Replicate it in your own niche, in your own packaging.

A niche is not a format either, and this is where almost everyone stalls. People test voiceover tools, optimise thumbnails, study analytics dashboards and research CPMs — all before answering the most basic question available: what is my format?

A topic gets someone through the door. The format is what makes them stay, and more importantly what makes the next video feel necessary rather than optional. Answer the format question first and every other decision — length, pacing, thumbnail style, upload rhythm — stops being a guess.

A $20 video that generated $10,498.98. That's a 52,394% ROI, and I did roughly 95% of the work myself. The reason it worked wasn't the budget — it's that the cost of being wrong was almost nothing, so I could afford to keep being wrong until something landed.

That's the part people miss when they see the ROI number and assume it was a fluke. The video wasn't brilliant. It was cheap enough that I could afford to make dozens like it, and one of them connected. Low production cost isn't a compromise on quality — it's what buys you enough attempts to find the winner.

1. A minimum 80/20 swipe rate as the floor.
2. Around 85/15 as the level where a big push became likely — almost every video hitting that got one eventually, even when APV was only around 70%.
3. Strong numbers on both swipe rate and APV within the first 48 hours, which reliably earned a second and third push over the following weeks.

The zero-CTA channel converted terribly and performed best. Around one subscriber every 2,000 views — but the best longevity and the most consistent real-time views of any channel I ran.

That result is the one I'd point at if you only take one thing from the whole experiment. The metric that looked worst on the dashboard belonged to the healthiest channel, and the metric that looked best belonged to the one that stalled. Optimising the number you can see is not the same as optimising the business.

1. Tone of voice and emotional tone.
2. Narrative structure — linear, mystery reveal, flashbacks, cliffhangers.
3. Pacing and sentence style.
4. Hook style — question, shocking fact, cinematic build.
5. Common devices: open loops, repetition, analogies, suspense.
6. Point of view and how the audience gets addressed.

The tax bill arrives years later, when the money is gone. People ride the wave up on internet money, spend like the wave is permanent, and get destroyed by a bill they never planned for. This is the single most common way I've seen people lose everything they made.

This is the failure mode I've watched destroy more people in this space than any algorithm update or demonetisation wave combined. The revenue feels like income because it lands in your account every month, so it gets spent like income. Then the bill for two years of it arrives at once, against a balance that's already gone.

1. Purpose. You don't know what problem you're solving, so your mind defaults to society's script and you drag your feet through a life you didn't choose.
2. Environment. Surrounded by people, opinions, noise and digital junk you control but never remove.
3. Metabolism. Big meals push you into rest-and-digest mode, so energy goes to digestion instead of output. Experiment with smaller meals, lighter mornings, eating later.
4. Knowledge. Sometimes you're stuck because you genuinely lack the skill. Bake learning into the work — audiobooks on walks, lectures instead of scrolling.

People worship consistency because it sounds mature. It sounds responsible. It sounds like advice from someone who has never done anything difficult. And if you are only consistent, you're not moving forward — you're staying busy, and stagnation isn't neutral.

And it is genuinely good advice for holding onto something you've already built. It's terrible advice for building it. If you're doing the same volume at the same quality every week for a year, you haven't been consistent — you've been stationary, and calling it discipline just makes it harder to notice.