Aakash Gupta@aakashgupta
MrBeast keeps going on podcasts and keeps giving away the entire YouTube playbook. Here’s what he’s said across dozens of appearances.
On the algorithm: it doesn’t exist. Replace “algorithm” with “audience” every time. The algorithm didn’t like your video. No. The audience didn’t. YouTube is a mirror. If people click and watch, it gets promoted. The growth hack industry sells you a god that isn’t there.
On what actually matters: studying humans. The checklist before you hit record. What’s the thumbnail. What’s the title. What’s the first 5 seconds. What’s the first 30. If you can’t answer all four, don’t film.
On titles: under 50 characters. Above that, devices cut them with dot-dot-dot and viewers don’t know what they clicked. Short, simple, so interesting it’ll haunt them if they don’t click.
On thumbnails: simple enough a scrolling viewer instantly understands and feels emotion. His test: “I rode a skateboard with 1,000 other people, it’s about to go off a big ramp.” Hours later, daydreaming, you still wonder what happened to those 1,000 people.
On autoplay: videos autoplay now. Many people never see the thumbnail. You have to visually convince them in the first 5 seconds.
On extremity: “Fiji water sucks” does fine. “Fiji water is the worst water I’ve ever drunk in my life” does way better. The more extreme the promise, the more extreme the delivery has to be.
On matching expectations: title and thumbnail set the promise. The first 10 seconds honor it or break it. Click “Tether is a scam” and the creator starts on anything else, you’re out. Start with “Tether is a scam and I’m gonna teach you why.” Match, then exceed. The thing people undervalue most is literally the first 10 seconds.
On retention: remove every dull moment. Find 10 critical people, make them watch, let them roast it. Ten seconds of talking head without a cut loses people. B-cam three seconds in, different angle, now it’s interesting.
On drop-off: creators drag it out. “I’m going to eat $100 ice cream, but first…” and then it’s them birthday shopping for their mom. Give them why they clicked. Tell them why to watch. Stay on topic. Upper echelon of YouTube.
On the real metric: it’s the next video. If they loved what they just watched, they watch your next one. You don’t want “that was good, but enough for the day.” You want “holy crap, what’s that?” and they watch 10 in a row.
On quality vs quantity: easier to get 5M views on one video than 50K on 100. Small creators post stuff that isn’t bad but isn’t great, nothing pops off, no audience forms. Upload a third or a fifth as often and make each one so good the algorithm has to promote it.
On the consistency trap: a schedule you can’t hit at quality is dangerous. “Monday I said I’d upload” floors your quality at exactly the level viewers notice. They watch less. Longevity suffers.
On the first 100: they’re going to suck. You think they’re good. They’re not. When he was 14 he thought his videos were the best in the world. They were terrible. Under 1,000 subscribers, your videos probably aren’t good yet.
On the improvement loop: ship 100, improve one thing each time. Second, better script. Third, new editing trick. Fourth, vocal inflections. Fifth, thumbnail. Sixth, title. No such thing as a perfect video.
On analysis paralysis: planning your first video for three months is the worst move. Your first 10 get zero views. Confirmed. Stop thinking, start shipping. On your 101st we’ll talk.
On the ceiling: “I could start a new channel tomorrow without my face, my voice, or promoting it, and hit 20M subscribers in six months. If you knew what I knew, you could get 10M from wherever you are.”
Every creator watching a 30-second clip thinks they got the tip. They got one tile from a mosaic he’s built in public for years.