ProvenStartups
IdeasPricingMethod
Get access
ProvenStartups

Startup ideas with revenue receipts, reverse-engineered from founder interviews.

contact@provenstartups.com
Product
  • All ideas
  • Pricing
  • Method
  • Blog
Company
  • About
  • FAQ
  • Contact
Legal
  • Privacy
  • Terms
  • Refunds
Official accounts
  • LinkedIn
  • GitHub
  • Reddit
  • Pinterest
© 2026 ProvenStartupsNo fabricated numbers. Ever.
Home/Blog/What It Really Pays

Highest Profit Margin Businesses: Where the Margin Really Sits

The highest-margin businesses in our index are software products that never touch inventory — app portfolios, AI tools and SEO products — and the top of…

ProvenStartups·Published 2026-08-01

The highest-margin businesses in our index are software products that never touch inventory — app portfolios, AI tools and SEO products — and the top of that list runs to $2.2M/mo. But across all 406 businesses we have indexed, only 106 disclose a clean monthly figure at all, and the median of those is $27K/mo, with a full range from $6/mo to $2.2M/mo.

That gap between the headline and the median is the entire subject of this page.

Table of contents

  • ·Which businesses have the highest profit margins?
  • ·The ranking, and the evidence behind each entry
  • ·How much is each of these numbers actually worth?
  • ·Why the usual rankings are wrong
  • ·What sits in the middle of this ranking
  • ·How to use this without fooling yourself
  • ·Frequently asked questions

Which businesses have the highest profit margins?

We sort by the strength of the evidence behind the figure, not by how good the idea sounds. Ours is stated and checkable — see how we grade the evidence, then disagree with us on the record.

The honest caveat, up front: this index over-represents businesses that both survived and chose to talk about it. Nobody films a video about the store that closed. Treat $27K/mo as a comparison baseline between categories, not a number you should expect to reach.

Close-up of hand pointing at finance report with charts and graphs on paper.
Photo by https://kaboompics.com/ on Pexels

The ranking, and the evidence behind each entry

The top of the index by disclosed monthly revenue, evidence class on every row. The class is not decoration — it changes what the number means.

BusinessWhat it earnsEvidence classLink
App Portfolio Studio ModelPeak $2.2M/mo (15 apps combined)Founder-reported/projects/app-studio
The Viral App Monetization MachineCal AI & Lerna $2M/mo each; LazyFit, CoinSnap & Impulse $700K/mo eachVerified, hard data/projects/paywall-machine
Retake AI Selfie Editor~$2M/mo (creator citing Sensor Tower)Creator-relayed/projects/retake-ai-selfie-editor
AI Clone of Floe-Style Agency SaaS$1.3M/mo for the original benchmark; the cloner himself: $0Creator-relayed/projects/clone-floe-saas-codeguide
OutrankPushing toward $1M/moFounder-reported/projects/outrank
Cluely$500K/mo (~$6M ARR)Founder-reported/projects/cluely-stealth-ai-assistant
Niche Identifier App Portfolio (CoinSnap)$500K/mo, top identifier app (Sensor Tower)Verified, hard data/projects/coinsnap-niche-id-app-empire
Quitter~$500K/mo (secondhand, via the Cluely interview)Creator-relayed/projects/quitter-vaping-quit-app
Gravl$440K/mo, 70K+ subscribersVerified, hard data/projects/gravl-ai-gym-workout-app
Leftclick~$400K/mo across businessesFounder-reported/projects/leftclick-automation-agency-skill-decay

Read the evidence column, not the revenue column. Retake AI Selfie Editor (Retake / AI Selfie Editor) at ~$2M/mo and Gravl at $440K/mo are not the same kind of claim, though one number is four times the other.

Note what the AI Clone of Floe-Style Agency SaaS row says: the original benchmark does $1.3M/mo, and the person demonstrating how to clone it made $0. Most rankings publish the first and drop the second.

How much is each of these numbers actually worth?

Across all 406 indexed businesses: 57 third-party verified, 184 founder-reported, 121 creator-relayed, 44 unproven. That is 59% verified or founder-reported; the remaining 41% is somebody repeating something they heard.

Evidence classCountWhat it means
Third-party verified57A source outside the business — app-store analytics, a platform dashboard, a public filing — supports the figure
Founder-reported184The operator stated the number themselves. Usually directional, rarely audited, sometimes revenue dressed as profit
Creator-relayed121Someone else repeated the figure. Fine as a lead, useless as a benchmark
Unproven44Claimed, with nothing attached

This is the argument the whole site rests on: a founder-reported figure and a verified figure are not the same claim, and quoting them in the same sentence without labels is a small act of dishonesty. Cleo claimed $60K MRR in 53 days with zero proof, and the video making the claim was itself a link in the launch funnel — marketing, not evidence.

Public statistics have the same problem in reverse. The SBA Office of Advocacy small business profiles and the U.S. Census Annual Business Survey are methodologically solid but aggregate — they tell you how many firms exist, not what one app portfolio grossed last month.

Overhead view of hands highlighting financial documents on a desk.
Photo by RDNE Stock project on Pexels

Why the usual rankings are wrong

Most such lists are sorted by search volume or by what the writer found exciting that afternoon. That is why the same eight categories appear in the same order across dozens of sites, none citing a single operating business by name.

Sorting by evidence changes the order visibly. Three effects:

  1. 1.Verified mid-size businesses climb. The Viral App Monetization Machine (100 Apps Dissected) and CoinSnap's $500K/mo survive because Sensor Tower data backs them. A larger unproven claim ranks below them.
  2. 2.Relayed giants get demoted. Quitter's ~$500K/mo is secondhand, repeated inside another founder's interview.
  3. 3.Failures stay visible. We keep 38 documented failures inside the same index rather than filtering them out, because a category's margin story is incomplete without them.

Two failures worth naming. Chai reached $30M/yr with 21 people and a $450M valuation — but its cold start was posting the product online to crowdfund a GPT-3 bill it could not pay, and 80K sign-ups in a day crashed the servers. The NoFap / Men's Self-Improvement App did $6K/mo in its first month with ~1,100 paying users — a real business, nowhere near the headlines above. Both belong in an honest ranking.

What sits in the middle of this ranking

The median disclosing business earns $27K/mo, not $2.2M/mo. That is the number to plan against, and 246 of the 406 indexed businesses are run by one person.

At $27K/mo you are typically looking at a single product with one distribution channel that works, a founder doing support and marketing personally, and no team. It is the Rook · Turnkey AI-Avatar Finance Channel profile — $47K over two months, peaking above $6,000 in a single day — rather than the Cluely profile with ~13-14 people and 60 contracted creators.

Margin at this level is a function of what you refuse to buy: no inventory, no office, no paid team. Services hold that line differently from software, which we break down in AI Automation Agency: The n8n/Zapier Margin Reality and How to Build an n8n AI Agent Business That Keeps Its Margin.

Close-up of hands holding and analyzing finance charts with pen.
Photo by https://kaboompics.com/ on Pexels

How to use this without fooling yourself

  1. 1.Pick your comparison number before you research. Use $27K/mo, the median of the 106 disclosing businesses — not $2.2M/mo, one founder-reported peak across 15 apps.
  2. 2.Label every figure you collect. Write "verified", "founder-reported", "creator-relayed" or "unproven" beside it. If you cannot assign a class, you have a rumour.
  3. 3.Read the failures in your category first. All 38 documented failures sit in the same index as the wins — start at all indexed ideas and read the losses before the highlights.
  4. 4.Sanity-check against a public baseline. Run your assumptions past the SBA's guide to planning a business, then compare category numbers in the Revenue Reality hub.

Frequently asked questions

What business has a 90% success rate?

None that we can document. Our index holds 406 businesses and 38 documented failures, but that ratio reflects who chose to disclose, not survival odds. Any site quoting a 90% success rate is quoting an unproven figure, the weakest of our four evidence classes.

What business can make $10,000 a month?

$10K/mo sits below our median of $27K/mo, so a large share of the 106 disclosing businesses clear it. The range starts at $6/mo, so nothing about the category guarantees it. Solo operation is common: 246 of the 406 indexed businesses are run by one person.

What percentage of businesses make $500,000 a year?

We cannot answer as a percentage of all businesses, only of ours. Of the 106 disclosing a clean monthly figure, the median is $27K/mo — roughly $324K/yr — so about half clear that pace. Cluely at $500K/mo and CoinSnap's portfolio at $500K/mo sit far above it.

How reliable are these highest profit margin business figures?

Partly. Across the index, 57 figures are third-party verified and 184 are founder-reported — 59% combined. The other 41% is 121 creator-relayed plus 44 unproven claims. We publish all four classes with labels rather than blending them into one average.

How many of the 406 indexed businesses actually disclose a monthly number?

  1. 1.The rest disclose something less useful — annual figures, ranges, growth claims, or nothing. Those 106 span $6/mo to $2.2M/mo across 22 countries, with a median of $27K/mo. We would rather publish 106 clean figures than 406 dirty ones.
← More in What It Really PaysBrowse proven ideas