"It's just a ChatGPT wrapper."
That's the insult, right? You ship something, some guy on X calls it a wrapper, and you're supposed to feel bad.
Meanwhile the three ChatGPT wrapper apps in this breakdown are pulling $20K, $30K, and $1,000,000 a month. Yes, one of them makes a million dollars every single month writing essays.
Pat Walls (the Starter Story founder) ran a teardown of all three on the Starter Story Build channel. He walks through each product, guesses the revenue from public traffic data, then reveals the real number from their founder interviews.
I pulled out the numbers, the math, and the lessons. Watch the original here, then let's break it down:
The 30-second formula for guessing any app's revenue
Before the apps, steal the method. Pat uses one napkin formula the whole video:
Monthly traffic × conversion rate × price = monthly revenue
Traffic comes from SimilarWeb (free tier is enough). Conversion rate, he assumes a conservative 0.5% of visitors become paying customers. Price is right there on the pricing page.
So a site with 100,000 visits, a 0.5% conversion rate, and a $99 product is doing roughly $49,500 a month. That's the whole trick.
It's a blunt instrument. Pat's guesses in this video land anywhere from half the real number to 2.5x over. But blunt is fine for deciding whether a market is worth entering. Run it on any competitor before you build.
1. Sheets Resume: $20K/month from a $99 resume builder
Sheetsresume.com is an AI resume builder. You paste your LinkedIn, it pulls everything over, and the AI writes your bullet points. It walks you through what a good resume looks like while generating one for you.
The numbers:
- Traffic: 134K visits in February, 152K in March, 120K in April
- Pricing: $99, one time, lifetime access, money-back guarantee
- Pat's formula guess: around $50K/month
- Real number: $20K/month (about $30K in their best month)
The margins on this are absurd, by the way. The product is generated text, and text is the cheapest thing an LLM produces. Colin collects $99 and his API bill for that customer is measured in cents. Old-school resume consultants charged $200 an hour to do the same job slower.
Why one-time pricing instead of a subscription? Because nobody needs a resume every month. You need one now, badly, to land an interview. Paying $99 once to get a real shot at a job you want is an easy yes. A $15/month resume subscription would churn instantly and feel gross.
The founder story is the actual lesson here. Colin McIntosh spent years as a recruiter. He was already giving people free resume advice all day, and his free resume template had been floating around Reddit since 2018. The paid AI product came after the audience and after the expertise.
He knew the customer so well that the expansion wrote itself: AI cover letter maker, AI mock interviews, an AI job board. Same customer, same stressful month of their life, four more things to sell them.
So before you scroll another idea database, answer this: what do people already ask you for help with? (For Colin it was resumes. For you it might be something you don't even consider a skill.) That's your wrapper.
2. Cuppa AI: $30K MRR writing SEO content
Cuppa.ai generates SEO articles at scale. You feed it your knowledge base and brand voice, it produces content briefs and full articles, and it publishes straight into WordPress.
The numbers:
- Traffic: around 40K visits a month
- Pricing: $25/month entry plan, $60/month on the highlighted "popular" plan
- Pat's formula guess: roughly $18K in new MRR a month
- Real number: $30K MRR, up from zero about a year earlier
One detail Pat skips that makes this story better: Chris, the current owner, didn't build Cuppa. He bought it from an indie developer in 2023 for $9,000 and made the money back the first weekend of sales. $9K in, $30K a month out. (And people say you need to be technical.)
The "you can just use ChatGPT for that" crowd loves this category. And sure, you could prompt your way to one article. Now do 100 articles, in 100 separate chats, each one on-brand, then copy every single one into WordPress by hand.
That gap is the product. Anyone can call the same model Cuppa calls. People pay $60 a month for the part around it: the saved knowledge base, the batch output, the button that publishes 100 articles into WordPress while they do something else.
If you want to see what building on top of ChatGPT actually looks like in practice, I broke down a full build in how to build an app with ChatGPT.
3. Aithor: the $1M/month essay writer nobody talks about
This is the big one. Aithor.com is an AI writing assistant for academic work: essays, research papers, theses. It suggests ideas, structures outlines, and handles citations in APA, MLA, whatever your professor demands.
The numbers are stupid:
- Traffic: 1.2 to 1.4 million visits a month
- 62% of it from organic search, and the brand name alone gets searched 43,000 times a month
- Pricing: $12.99/month billed annually, $35 month-to-month
- A live counter on the pricing page claimed 1,490 people subscribed that day
- Pat's formula guess: about $195K in new MRR a month
- Real number: $1,000,000 a month
A million a month. From a writing tool for students. While everyone argues on X about which AI startup will win enterprise, a niche essay helper quietly crossed eight figures a year.
Notice the pricing spread: $12.99 annual vs $35 monthly. Students write a paper, cancel, come back next semester. Churn is brutal, so Aithor pushes hard on annual and prices the monthly plan for the people who'll leave anyway. They built the pricing page around how students actually behave instead of fighting it.
How did they get 1.3M visits? Influencers on TikTok and Instagram, plus word of mouth compounding into that 43K/month brand search volume. Students live on TikTok. Aithor went where the customer already was.
And the market is enormous. Think about how many people are sitting in a college class or a master's program right now, all of them required to write papers most of them hate writing.
What separates the $1M wrapper from the $20K wrapper
Three apps, same "wrapper" architecture, revenue spread of 50x. The difference isn't the tech. All three sit on the same models you have access to today.
What actually moved the number:
→ Niche beats broad by an order of magnitude. Cuppa serves "businesses that want content" and makes $30K/month. Aithor serves students writing academic papers and makes $1M/month. Pat's take, and mine: the niche product gets to dominate its category instead of competing with every AI writing tool on earth.
→ Each one solves a moment, and the moment sells. Help me get a job. Help me rank on Google. Help me survive finals week. The customer never once thinks about which model is under the hood.
→ Pricing follows the customer's rhythm. One-time need, one-time price (Sheets Resume). Ongoing need, subscription (Cuppa). Spiky seasonal need, cheap annual plan to fight churn (Aithor).
→ All three founders started from a customer they already understood. The ex-recruiter built the resume tool. The SEO guy bought the SEO tool. That knowledge is the part nobody can rent from OpenAI.
If these numbers got you itching to build something small, I put together more micro SaaS examples with real revenue attached.
FAQ
Are ChatGPT wrapper apps actually profitable?
Yes. The three in this breakdown make $20K, $30K, and $1M a month. Margins are strong because generated text is cheap: Cuppa's output costs pennies per article, and Sheets Resume's cost per resume is close to zero. The profitable ones sell a workflow and a niche outcome, not raw AI access.
What counts as a ChatGPT wrapper?
A ChatGPT wrapper is an app whose core feature calls an AI model (OpenAI's or anyone's) through an API, wrapped in an interface built for one specific job. Sheets Resume wraps AI in a resume workflow. Aithor wraps it in academic writing with citations. The "wrapper" part is the model; the business is everything around it.
How do I estimate how much a web app makes?
Multiply monthly traffic (from SimilarWeb) by an assumed conversion rate (0.5% is a conservative default) by the price on their pricing page. A site with 100K visits and a $99 product lands around $50K/month by this math. Expect to be off by 2x in either direction; it's for sizing markets, not audits.
Won't OpenAI just kill these apps?
It's the standard objection, and these numbers argue against it. ChatGPT has existed the entire time all three apps grew. General-purpose chat doesn't ship your resume format, your WordPress integration, or your citation styles. The narrower the workflow, the safer the wrapper.
Want the full playbooks, not the guesses?
Every week on the Profitable Founder Podcast I sit down with bootstrapped founders doing $100K to $10M a year and get the real numbers out of them: pricing, channels, what flopped.
