Micro-SaaS listings are priced on revenue and bought on faith. The work here is turning the first number into the one the professional market actually uses.
Marketplaces connect to Stripe and confirm the money arrived. That is useful and it is not diligence: nothing in a verified revenue figure tells you the cost base, the churn, the concentration or the owner's hours — and every one of those changes what the business is worth to you.
01
Annual prepayments booked at collection make a single month look like growth. A price rise shortly before listing does the same thing for a quarter. The fix is the revenue curve across 24 months and the archived pricing page, not the headline number.
02
Logo churn and revenue churn answer different questions and a listing gives you neither. At 8% monthly logo churn the average customer is gone inside a year, which changes what the business is worth far more than the current MRR does.
03
If one account is more than a fifth of revenue, you are buying a relationship you have not met. Concentration is a valuation haircut in every professional process and a footnote in almost every listing.
04
Support, sales, onboarding and the occasional deploy add up to a job. Price the hours at what an operator would cost, subtract that, and see what is left. Profitable at acquisition, not profitable once you fix it.
05
Domain age and backlink quality predict revenue in this market better than anything about the product. A business ranking on a decade-old domain and one running on the founder’s audience are different assets at the same MRR.
06
A thin wrapper over a model API has a dependency that is not on the balance sheet. It matters what the margin looks like when inference prices move and what remains defensible if the capability arrives natively.
A listing at two times revenue could be four times profit or forty. The gap between those two sentences is the entire risk in a micro-acquisition, and closing it is half an hour of work for someone who does it professionally and effectively impossible from a listing page.
Strip out hosting, model APIs, tooling, payment fees, support and contractors. Normalise the founder's labour to what replacing it costs. What remains is the number you are actually buying, and it is the only one worth comparing to anything.