REEL #47 · THE FULL BREAKDOWN

They made $1,000,000 from an app that makes you do push-ups.

Two people. Two screens. The first version took two weeks. The thing that counts your push-ups through the camera is a free Google library anyone can install today.

This page is the playbook from the reel: how to find an idea like this, how to build it, and the marketing system that decides whether you make a million or six thousand. Every number links to the place it came from.

See the app → The founder interview Build it yourself

$1,000,000+ in a year300,000 downloads2 people2 weeks to v1$0 on ads

What it actually is

You pick the apps you want blocked. When your time runs out, Instagram will not open until you do push-ups. Your phone camera watches you and counts the reps, so you cannot cheat it. Ten push-ups buys you ten minutes.

The Pushscroll block screen: Instagram is Blocked by Pushscroll — The block screen. Their own App Store screenshot.

The Pushscroll home screen with exercise tiles — Two screens. Pick an exercise, earn minutes.

Source: the app's own App Store listing, read 2026-09-18. 4.7 stars from 18,000 ratings, seller Pushscroll GmbH.

The part everyone skips: they faked it first

Before they wrote a single line of code, one of them filmed a video and edited it to look like the app already existed. It did not. He posted it anyway.

That video got 80,000 views and a comment section full of people asking where to download something that was not real.

“I basically made a fake product demo and it got 80K views.”

Alejandro Sanchez, on the Starter Story interview

So the video was not marketing. It was the market test. They only built the thing after the comments proved people wanted it.

The lesson, in one line: the cheapest way to find out if anyone wants your app is to post the app before you build it. A video costs an afternoon. A build costs two months.

Then six months of almost nothing

This is the part that gets cut out of every success story. The fake demo went viral, they built the app, and then almost nobody used it for half a year.

What finally worked was realising the videos did not need to be about the app, or even about fitness. They needed a ridiculous visual hook: someone doing push-ups somewhere you would never expect. Two in the morning. In the rain. Then the app.

His actual TikTok grid. The same shot, over and over, 72.7K likes. That repetition is the product.

How you would build it

Nothing in this app is rare. Here is every piece, and who it actually belongs to.

🔒The blocker

A standard phone screen-time feature. Both platforms expose it to developers.

Not theirs

👤The rep counter

Google MediaPipe pose detection. Free, open source, runs on the device.

Not theirs

📱The app

Two screens. Pick the apps to block, do the reps. The first version took two weeks.

2 weeks

💰The paywall

One hard paywall, around $30 a year. No free tier to speak of.

Off the shelf

🎬The content formula

One video shape that worked, made over and over until it printed money.

This was theirs

Google's MediaPipe pose landmark detection documentation — Google's MediaPipe pose landmarker. This is the thing counting the push-ups, and it is free.

MediaPipe pose docs → Apple Family Controls Android usage stats

One honest note. Pushscroll's own sources never name which blocker API they used, so the two links above are how apps of this type are normally built, not a claim about their code. Their sources do not say the app was built with AI either.

The money, with receipts

$30Ka month in profit, four months after launchSuperwall, Oct 2025

$1M+total sales across both stores, in one yearBraavo, republishing the founder's own post, Jul 2026

28Mpush-ups done inside the appBraavo, Jul 2026

$0spent on ads to get thereSuperwall, Oct 2025

Superwall's write-up: Our App Makes $30K/Month Profit — Superwall, Oct 2025.

Braavo's write-up: From $0 to $1M — Braavo, Jul 2026.

Read this carefully. The $1,000,000 is total gross proceeds across both app stores, before Apple and Google take their cut, before tax, and before it is split between two people. It is not profit and it is not what either of them took home. The figure that is explicitly called profit is the $30K a month.

Superwall write-up Braavo write-up

Now the part that actually matters

The idea is not the moat. Other apps have the exact same lock, and they are not making this money.

Same idea, verified300,000users on Quran Unlock$13,252earned in totalTrustMRR, payment-provider verified, read 2026-09-18

A different app40,000,000views a month$6,000a month in revenueBraavo, Jul 2026. The app is not named in the source.

TrustMRR's verified page for Quran Unlock — Quran Unlock on TrustMRR. Same lock mechanic, 300,000 users, $13,252 all time.

Same mechanic. A fraction of the money. So the views were never the business, and neither was the code.

The playbook, step by step

Find an idea you can say in one line

“Do push-ups to unlock Instagram” needs no explanation and is instantly visual. If your idea needs a paragraph, it cannot be a video, and if it cannot be a video, you have no distribution.

Post it before you build it

Make the video of the app working. Edit it so it looks real. Post it. If the comments ask where to download it, you have a market. If they do not, you just saved yourself two months.

Build the smallest version that does the one thing

Two screens. No accounts, no feed, no settings nobody opens. Use the platform features and the free libraries. Two weeks, not two months.

Find one shot that works, then repeat it forever

Not ten ideas. One format, varied by location and situation. They did push-ups somewhere unexpected, then showed the app. That is the entire creative strategy.

Charge from day one

One hard paywall, around $30 a year in their case. 300,000 downloads with no paywall is a hobby, which is exactly what the comparison above shows.

The uncomfortable part. Building is the half that feels like work, because it has a finish line. Posting the same video for the hundredth time does not. That asymmetry is precisely why most finished apps have no users.

How every number here was checked

Each figure on this page came from a named source that I read directly, on the date shown. Nothing is estimated and nothing is from a revenue-guessing site.

$30K a month in profit, $0 on ads — Superwall's write-up, Oct 2025.
$1,000,000+ total proceeds, 28 million push-ups — Braavo, Jul 2026, republishing the founder's own post. Corroborated independently by Pat Walls of Starter Story, who interviewed them.
300,000 downloads, $30 a year, two founders — the founders on the Starter Story interview.
Quran Unlock: 300,000 users, $13,252 — TrustMRR, verified through the payment provider, read 2026-09-18.
4.7 stars, 18,000 ratings — the live App Store listing, read 2026-09-18.

Deliberately left off this page: any revenue estimate from a third-party analytics site, any claim about which blocker API Pushscroll uses, and any claim that the app was built with AI. No source I could open states those.