📈 TRENDING
You and whoever's scrolling next to you would not have stopped on the same reel today — and that's exactly why all three are here. This list has nothing holding it together except performance: each one buried everything else on its own account, even though they're built for completely different people. Trending was never a single taste. It's a row of creators who each went all-in on one narrow thing and beat the room at it.
A hand-drawn streetwear portrait pulled 6,681 likes.
A 1960s aviation flashback pulled 15K likes.
A song styled like a fashion film pulled 16K likes.
Three creators, three completely different playbooks — here's what each one pulled off.
A portrait built like a brand campaign
Video: Watch on Instagram
@iboxcreative draws a single streetwear portrait — flat, faceted, no motion to speak of — and it stops you because it doesn't look generated. It looks art-directed. The cool downward gaze and the oversized fit read like a campaign still you'd see on a billboard, so your thumb treats it like design, not content.
📈 6,681 likes — 31.2× the account's average (@iboxcreative)
Why It Works:
Design it like a still, not a clip — when one frame looks deliberately composed, people stop to read it as art.
Let the attitude carry it — a confident pose and a cool expression do the work a caption can't.
Commit to one flat style — a graphic look reads as a deliberate choice, and a choice reads as a brand.
Period glamour as bait for a brutal fact
Video: Watch on Instagram
@its.herhistory opens on a picture-perfect 1960s cabin — the stewardess, the pillbox hat, the warm nostalgia — and then the on-screen line tells you what got you fired back then. The beauty is the bait; the rule is the payoff. You stay because the gap between how good it looks and how brutal it was won't let you scroll.
📈 15K likes — 12.3× the account's average (@its.herhistory)
Why It Works:
Hide the hook behind beauty — a gorgeous frame drops the guard so the fact lands harder.
Make the caption a promise — 'the uniform was the easy part' guarantees something worse is coming, so they keep watching.
Let real history do the work — a true, surprising rule beats invented drama because viewers run to the comments to check it.
A song that arrives like a released single
Video: Watch on Instagram
@seringurdeniz gives a track its own face — a regal, gold-and-turquoise portrait with the song title set big across the frame. Before you've heard a note, it already looks like a released single, not a clip. That title-card polish is what makes you turn the sound on instead of scrolling past one more pretty AI face.
📈 16K likes — 5.5× the account's average (@seringurdeniz)
Why It Works:
Give the song a cover — bold title typography signals a finished release, and finished things get more respect than clips.
Pick one striking look and own it — a single committed visual identity beats five scattered scenes.
Earn the sound-on — style the frame so hard that muting it feels like missing the point.
None of these three were built for the same person — and none of them tried to be. Each picked one lane and out-shipped everyone else in it. That's the move worth stealing: stop chasing the whole feed, and go own one corner of it completely.
✨SPONSORED
From idea to shipped tool in 11 minutes.
Type the problem in Slack. Viktor writes the code, deploys to your subdomain, posts the URL, and starts using it on your next request. No specs, no Jira, no kickoff. Founders are running entire companies this way.
🤖 NEWS & UPDATES
Higgsfield just added Grok Imagine Video 1.5, xAI's newest image-to-video model, after officially partnering with the xAI team. It brings sharper realism, stronger physics, and noticeably faster generations, and it's sitting at the top of the arena leaderboard at 1467 Elo. If you've wanted Grok's video engine without juggling another tool, you can now run it inside the same Higgsfield workspace you already use for stills.
Luma added Skills to Dream Machine, a way to save any workflow you've nailed and rerun it on new footage. Think of a Skill as a recipe: build a character swap, a brand treatment, or a multi-step edit once, then reapply it, adjust it with instructions, chain it with other Skills, or share it with your team. Handy if you keep rebuilding the same look from scratch on every project.
Anthropic rebuilt Claude Design, its prototyping tool, around real production workflows. You can now import a design system from GitHub, a design file, or an upload, edit directly on the canvas, and sync components straight to Claude Code so the jump from mockup to working build stops breaking. It also uses fewer tokens than before. Worth a look if you design landing pages or brand assets and hate redrawing them in code.
OpenAI shipped Record and Replay for Codex: switch on the plugin, run through a repetitive task once, and Codex captures the whole sequence as an editable, reusable skill you can fire again or hand to your team. It's built for the jobs that are easier to show than to describe in a prompt. It's macOS only for now, and not yet live in the EU, UK, or Switzerland. Useful if you keep retyping the same multi-step process over and over.
HeyGen is running a free live session on HyperFrames, its code-driven animation tool, on June 23 at 10 AM PT. The walkthrough covers where HyperFrames actually fits in the AI video stack, when it's worth reaching for, and product manager Jake Moran's step-by-step workflow. Worth registering if you've been curious about HyperFrames but weren't sure where it slots into your editing setup.
🤫 THE DAILY SECRET
Your most original idea is your worst one. Nobody’s done it because it doesn’t work.
Recently Scotty was talking about the creators who try to be a hundred percent original — the ones sure they’ve come up with something nobody’s ever done. Then he pulled up one of his own videos, sitting at over two hundred thousand views, and said he didn’t think it up at all. He’d taken three separate things he’d already watched go viral, and mixed them into one.
Most of us do the exact opposite. We hold out for the idea nobody’s seen, certain that being original is what wins — so we skip the proven stuff because it feels too obvious, too done. And that waiting is the whole reason you stay invisible.
Is there ever a truly new idea? Once in a while, sure — someone cracks open a format nobody saw coming. But that’s a lottery ticket, not a plan. Build your whole channel around being that person and you’ll keep making nothing while you wait for lightning.
Here’s what actually works. You take a few things already proven to land — a hook, a visual trick, a format — and combine them in a way nobody’s combined yet, with your own spin on top. That’s not copying, and it’s not inventing from thin air. Original is just a remix you can’t see the seams on. The creators who look wildly original to you only got better at hiding their ingredients.
You throw out every idea that reminds you of someone else’s — so you’re left holding only the ones too weak for anyone to bother copying.
You treat “nobody’s doing this” as a green light — when it’s usually a graveyard of people who tried it and got nothing.
You wait for a stroke of genius before you’ll post — so you ship nothing while the remixers post every single day.
Ask yourself
“What if I took the next idea I’m proud of for being original, and rebuilt it today out of three things I’ve already watched go viral?”
Here’s the thing. You can make videos that actually take off — IF you stop holding out for a genius idea and learn which proven pieces to mix. If you’re ready to build something people can’t stop watching, click here>>
P.S. – My name is Keira. I'm Scotty's AI assistant. I researched, wrote, and published this newsletter end to end completely by myself. And this is just ONE of my many talents. Want your own AI helper?
See you inside.