An app built in 12 hours now makes $15,000 a month.
Another one, prototyped over a single weekend in Swift, also sits around $15,000 a month.
And a job board thrown together in about 30 hours? A million dollars a year.
I know how that sounds. Every second YouTube thumbnail promises you a $10K/month app by Sunday night. Almost all of them are selling a course.
This one is different. Pat Walls from Starter Story pulled three real businesses from his case study database and walked through how each got built and launched, with launch posts you can still go read today.
So can you actually build an app in a weekend that grows into $10K a month? Watch the video, then stick around, because what made these work was not the code.
AudioPen: 12 hours of work, $15,000 a month
AudioPen is a dead simple idea. You talk into it, it turns your messy voice note into clean, structured text.
That's it. That's the whole app.
Louis Pereira built the first version during a hackathon called the Half Day Build. The whole point of the event: go from idea to revenue in 12 hours.
His own words from the case study: he was "forced to build the simplest version of the product that people might pay for." He did. And people paid.
Forced is the key word.
Give yourself three months to build an app and you'll take three months. Give yourself 12 hours and suddenly every feature that isn't core gets cut. Parkinson's law, working in your favor for once.
Two things Louis did that most weekend builders skip:
→ He wireframed the app in Figma before writing a line of code. Every screen, the pricing page, the whole flow. Designing first forced the UX to be good before anything got built.
→ He live-tweeted the entire build. One of his build-in-public posts pulled 115 likes and 25,000 views. Those views became his first users, and his first users became paying customers.
Also worth saying: Louis is building from Goa, India. $15K a month is life-changing money almost anywhere. There, it's a different life entirely. Software lets you serve customers worldwide from wherever you already are.
BoltAI: a weekend Swift prototype on a laptop
Daniel Nguyen wanted every AI model in one native Mac app. Claude, GPT, Grok, all of them, one keyboard shortcut away.
He had a product engineering background, so he built the first version in Swift over a weekend. It was ugly. He shipped it anyway.
Then he posted it. The launch post was one line about having every AI model in one app, plus a short demo video. It did 251,000 views, 833 likes, and 63 comments.
That single post seeded a user base most founders spend a year of paid ads trying to buy.
At the time of his Starter Story case study, BoltAI was doing $15,000 a month. Today his site gets 100,000 to 150,000 visits a month, and he keeps a public counter on his page tracking his progress toward a million. He's about halfway.
Notice the pattern repeating: rough prototype, shipped fast, shared publicly. The polish came later, funded by revenue.
One more idea from the video that I keep thinking about. A general "all models in one app" tool is good. The same tool hyperfocused on one vertical might be better. Pat mentioned a founder doing $20,000 a month with an AI app built for one single profession. One niche with one painful workflow, and nobody else building for it.
The weekend build is the easy part. Picking the niche is where the real work hides.
4 Day Week: a 30-hour job board making $1M a year
This one is my favorite, because it's the least technical of the three.
4 Day Week is a job board for companies that run four-day work weeks. Built in about 30 hours. Now doing $83,300 a month. That's a million-dollar-a-year business, run lean.
Here's the launch sequence, step by step:
→ Simple landing page in Webflow. No product behind it.
→ Posted it on Hacker News as a Show HN.
→ 400 signups in the first two days.
→ Only then did he build the real version in React.
Go find that Hacker News post. It has 70 upvotes and a handful of comments. There was no viral moment and no press coverage. Seventy upvotes changed this guy's life.
The business model is what makes the math beautiful. Companies pay to post jobs, starting at $199 a month. Pat ran the numbers on camera: at roughly $300 average per post, a million a year is about 276 job posts. Nine new posts a day.
You don't need 10,000 customers or a sales team for that. Just nine companies a day who are desperate to stand out in hiring.
Job boards are boring, and boring is underrated. I wrote about this exact dynamic in simple apps that make money: one feature, one audience, priced for businesses instead of consumers.
The real pattern: none of these made money in a weekend
Let me kill the fantasy version of this before you copy the wrong lesson.
Nobody built an app on Saturday and woke up to $10K MRR on Monday. Pat says this straight out in the video.
What every one of these founders did in a weekend:
→ Built the simplest version someone might pay for.
→ Launched it that same weekend, ugly and unfinished but live.
→ Shared the process publicly. Twitter, Hacker News, wherever their people were.
The money came months later, from the feedback loop the launch started. Users showed up, complained, requested, churned. The founders iterated. That loop is what grows into $15K a month, and you cannot start the loop until something is live.
Most builders get this exactly backwards. They build for six months, add features nobody asked for, and never launch. Until something is live, you're not running a business. You're rehearsing.
I've been there. Building in the dark feels productive. It's mostly hiding.
If the sharing part feels cringe to you, read how to build in public. It's the highest-leverage marketing available to a solo founder with no budget, and these three case studies are the proof.
How to run your own weekend build
Steal the playbook directly from these three:
Step 1: Pick a painful problem, not a cool idea. Pat's filter for every business he studies: what painful problem does this solve? AudioPen solves "my voice notes are a mess." 4 Day Week solves "hiring is expensive and we need to stand out." If you can't name the pain in one sentence, pick again.
Step 2: Wireframe before you build. An hour in Figma sketching every screen. It feels like a detour. It's the opposite: you'll cut half your features before they cost you a weekend, and you can feed the wireframes straight into AI tools to generate the actual pages.
Step 3: Set a deadline that forces cuts. Louis had 12 hours because the hackathon said so. You don't need a hackathon, you need a hard stop. Tell people you're launching Sunday night. Now Parkinson's law works for you.
Step 4: Launch before you're comfortable. No pricing page needed. A landing page and a signup form validated a million-dollar business for 4 Day Week. Post it on Reddit, X, or Hacker News and watch what happens.
Step 5: Share the whole thing. Every one of these founders got their first users from a public post, not from ads. Build something this weekend, show your work, and you genuinely cannot lose. Worst case, you learned. Best case, 70 upvotes changes your life.
FAQ
Can you really build an app in a weekend?
A simple, launchable version, yes. AudioPen went from idea to paying customers in 12 hours, BoltAI's first Swift prototype took one weekend, and 4 Day Week's validated landing page took even less. What you can't do in a weekend is build the polished version those products are today. You build the seed, launch it, and let user feedback grow it.
Do I need to know how to code?
Less than ever. Louis built the first AudioPen with Bubble and the OpenAI API, no traditional stack. 4 Day Week started as a Webflow page. With AI coding tools on top, a non-technical founder can get a working prototype live in a weekend. The bottleneck is choosing the right problem, not the code.
How do weekend apps get their first users?
All three of these got users the same way: one public post in the right place. A build-in-public tweet (25,000 views), a launch post (251,000 views), a Show HN (400 signups in two days). Free distribution from showing your work beats a paid ads budget you don't have.
How long until a weekend build makes real money?
Months, not days. The weekend gets you a live product and your first feedback loop. The revenue comes from iterating on what real users tell you. Every founder in the video launched fast, then spent months compounding: that's how a 12-hour build ends up at $15K a month.
Build it, launch it, then tell me about it
The gap between people who have an app making money and people who have an idea is almost never talent.
It's that one group launched an embarrassing v1 on a Sunday night and the other group is still polishing.
Pick a real problem, block out one weekend, and ship the ugly version.
And if you want to hear founders a few steps ahead of you break down how they turned scrappy launches into real revenue, that's literally every episode of my podcast.
