Micro SaaS Examples With Revenue: 93 Cases, Graded

ProvenStartups indexes 93 SaaS cases. Twenty-seven are micro SaaS under a strict reading — one person or a team of five or fewer, plus a disclosed monthly figure — and they run from $598/mo to $600K+/mo, median $16K/mo. The ceiling is not the story: fifteen of the 27 sit at or under $20K/mo, and only 13 of the 93 carry a number checked by an outsider.

Contents

• ·How much do micro SaaS businesses actually make?
• ·Which cases have the strongest revenue evidence?
• ·What separates the top cases from the bottom?
• ·What are the failure modes?
• ·Who should start one, and who should not?
• ·Frequently asked questions

How much do micro SaaS businesses actually make?

The median is $16K/mo and the range is $598/mo to $600K+/mo. Of the 27 qualifying cases, six sit under $10K/mo, fifteen at or under $20K/mo, and only three clear $100K/mo. Drop the single $600K+ outlier and the median moves to $15K/mo — this distribution is flat, not top-heavy.

How we selected: from the 93 records we kept only cases with five or fewer people and a clean monthly figure. That drops Kovai.co (~250 staff) and Buildern (35 full-time), plus the ten records with no figure at all — SpecSheet and PermitSync among them, price lists with no customers.

Grades matter more than size. ✅ is third-party verified, 🗣 founder-reported, 📎 relayed by a creator who never audited it, 🔮 unproven. All 93: 13 ✅, 53 🗣, 18 📎, 9 🔮. The 27: 8 ✅, 15 🗣, 4 📎. Two thirds of what follows is the operator's own claim.

Which cases have the strongest revenue evidence?

Nine of the 13 verified cases are small-team, and these 15 rows lead with them. Order is evidence grade first, index rank second — never revenue, because a verified $11,000/mo is a better thing to copy than an unaudited seven-figure claim.

Read the top against the bottom. Review Harvest's verified software line is ≈$36K MRR, but $32K of its $69K month is HighLevel affiliate revenue — hence its own rule that software is a sellable asset and commissions are not. Margin, not the headline, keeps two people in business.

What separates the top cases from the bottom?

Three things, in order: who you charge, where the customers come from, and how narrow the problem is. Team size and tooling barely register — the $96.2K MRR case and the $598 MRR case are both two-or-fewer-person operations shipping comparable software.

Does pricing explain the gap?

Mostly, because price sets how many customers you need. Hero Analytics sells to marketing agencies instead of the brands they serve, so one sale onboards 30 brands and pricing scales with client count ($500 at 10 clients): two cofounders, ~$96.2K MRR. UGC Tank, a comparable solo build, sells to brands one at a time and sits at $598 MRR.

Deliberate under-pricing also works. A Japanese solo developer took a tool agencies sell at ¥100K/mo, priced it at ¥5,000/mo for sole traders, and reached ¥200K/mo at roughly 10% free-to-paid conversion. Accidental under-pricing is the commoner failure: True Horizon AI moved one deal from a guessed $1,650 to $12,000, and calls a close rate above 50% proof you are charging too little.

Does the channel explain it?

It explains the speed. XBoost broke ¥3M/mo in its first month on zero ad spend, off one launch post that did 13.5 million impressions. SuperX reached $13,000 MRR at $29/mo with 95% of traffic from the founder's own X account — after five products that made exactly $0. BridgeMind credits streaming for every paying customer; Algrow chose Discord because nobody was fighting over it.

Note what is absent: paid acquisition. Not one of the 27 credits ads as its main engine.

Does niche depth explain it?

It explains durability. Packager automates an hour of Microsoft Intune deployment tedium and does $60K/mo, verified, because the job is too boring to attract copycats. WeeNote earns ₩20M/mo from under 100 Korean schools. A Dutch founder replaced his consulting income at roughly 10K/mo issuing road-exemption permits — and never says which currency, which is exactly what keeps a figure unverified.

What are the failure modes?

Four, and each is visible before you write code: a free tier that never converts, unit costs above the price, copying a product without its distribution, and buying growth you cannot show a buyer. The SaaS metrics hub defines the ratios.

1. 1.A free tier that never converts. Unified has about 500 paying users out of 11,000 — 1.5–2% — and never states revenue. Tally makes it work at roughly 2% of 800,000 users, a scale almost nobody reaches. Shipyard runs the opposite policy: ~690 paying users, zero free users, $25.6K MRR.
2. 2.Unit costs above the price. Rhythm.ai hit $2,500+ MRR in under 40 days and stayed near break-even, because its heaviest 10% of users each ran about $67 in the red at ~11 cents per minute of AI cost.
3. 3.Copying the product, not the distribution. A creator cloned an agency SaaS benchmarked at $1.3M/mo. The cloner's own revenue: $0.
4. 4.Buying growth you cannot show a buyer. Kaleb's first SaaS reached $10–11K MRR on $40,000+ of lifetime ad spend, then missed a $112,000 sale partly because the traffic came from twelve burner TikTok accounts rather than an ad account.

A fifth is for readers rather than founders: figures that do not reconcile. CodeTree sold for $128,000 cash at 3.5× SDE, yet its transcript puts peak MRR at "just over $44,000" — impossible at that price. We flag it instead of repeating it.

Who should start one, and who should not?

Start one if you already reach a specific group of buyers who pay for a boring, repeated task. Do not start one if the plan is a free tier plus a general audience; that shape stalls at four figures more often than any other here. meet.bot drew 12 signups on announcement day.

The cheapest honest test here is the pre-sale. Subscribr collected $20K from 50 lifetime licences before the product existed. MeetOscar ran a $5 skip-the-line deposit against 200 landing pages and was profitable from day one, at $45,000 MRR after 60 days. Both treat a credit card, not a waitlist, as the signal — and marketplaces like Empire Flippers are where that discipline eventually gets priced.

More in SaaS metrics — start with what micro SaaS actually is, then SaaS examples with revenue and apps built with Claude Code.

Frequently asked questions

What counts as a micro SaaS?

For this page: a software subscription run by one person or a team of five or fewer, with a disclosed monthly figure. That rule cut 93 SaaS records to 27. It excludes 35-person and 250-person companies selling comparable products, and records with a price list but no customers.

How long does it take to reach $10K MRR?

Months to years, with no reliable median. The fastest case here reached $10,000 MRR in six weeks on zero marketing spend, relayed by a creator rather than verified. Another needed 10 months, and one two-person business took 19 months to pass $1M ARR. A three-month-old solo product in the same file sits at $716 MRR.

Can a non-technical founder do this?

Several of the strongest cases are. A verified $21.8K/mo product came from a non-technical solo founder on borrowed distribution, a $25.6K MRR builder from two non-technical brothers, a $45,000 MRR email assistant from a non-technical operator. The cautionary case is also non-technical, so the variable is distribution, not coding.

Does a ✅ grade mean a case is safe to copy?

No. ✅ means someone outside the business checked the figure. It says nothing about churn, margin, market size, or whether the channel is still open. Only 13 of the 93 SaaS records reach that bar. Use the grade to weight the number, then judge whether you can reach those customers.

What do micro SaaS businesses sell for?

The file offers comparables, not a rule. One solo SaaS doing $26,877/mo with $16,661/mo profit was listed at $883,024, about 53× monthly profit. A nights-and-weekends product sold for $128,000 cash at 3.5× SDE. A $10K MRR Reddit tool sold for six figures. One $10–11K MRR business could not close at $112,000.