Do the valuation rules of thumb hold up?

We are not attacking a straw man, so here is the rule in someone else's words. Flippa's valuation guide, updated 16 October 2025, states: “A commonly cited orientation range is 30–45× monthly profit (about 2.5–3.75× annual profit)”, and repeats it later as “A widely cited benchmark is 30 to 45× monthly net profit, which works out to 2.5 to 3.75× annual profit.” (source)

Read that citation for what it is: evidence that the rule is in circulation and how it is phrased, from a marketplace that says so on its own guide. It is not a measurement we have checked, and we are not treating it as one. What we can check is our own side — what the sales we hold a source URL for actually changed hands at.

Loading live medians from /api/stats…

A category with fewer than five priced comps publishes no median at all rather than a thin one — see how many sold comps a valuation needs. Multiplying an annual multiple by twelve is arithmetic, not a second measurement: it restates the same median in the units the rule is quoted in.

A gap that large is not a rounding error, and it is not evidence that one side is lying. It is what happens when two numbers describe different populations, different denominators and different events.

1. Different businesses. The rule describes a screened listing: a profitable site with clean books that a marketplace or broker agreed to represent. Our comps are the sales that end up published with a price attached, and at this size that skews heavily toward auction closes at the very bottom of the market — median sold prices in the hundreds, not the hundreds of thousands. See what small online businesses sell for for the price column itself, and where sold-price data comes from for why the published sample looks the way it does.
2. Different denominators. A broker's multiple sits on a seller's discretionary earnings figure that has been through diligence. Ours sits on whatever profit the source page stated, unaudited — and an overstated profit figure produces an understated multiple on the same price. Several of our medians below 1× are that arithmetic happening in public; the SDE page takes the denominator apart.
3. Different events. A rule of thumb is mostly used to set a price — it prices asks, and asks are not sales. We only record sales, and the gap between the two runs in one direction at this end of the market. That gap has its own page: asking price vs sold price.

The conclusion we are not drawing is that the rule is a fabrication. A convention can describe one tier of a market accurately and say nothing whatsoever about another, and our data has no standing to referee a tier it cannot see: most brokered sales close with no public price at all, which is why so few of them reach any comp set, ours included.

What our medians are evidence about is the market that publishes its outcomes. If that is the market you are transacting in — a small auction close, a sub-$10k site, a first sale to a stranger — then a 30–45× ask is being quoted at you from a different tier, and the comps are the better guide. If you are selling a screened, verified, six-figure business through a broker, our sample is not describing your deal, and we would rather say so than let a median built from micro-auctions be read as a ceiling on your price.