The cheaper kitchen business on Flippa earned 2.84 times the profit.

Start with the pair, because they are as close to a controlled experiment as this market produces. Flippa publishes a quarterly card for each closed deal: asset type, sale price, annual profit, broker, buyer location, days to close. In the Q1 2026 set, Kitchen Ware Ecommerce sold for $425,000 on $851,412 of annual profit and revenue “exceeding $4.8M,” brokered by Daniel Kenny to a London buyer in 63 days. Kitchen Blog, “founded in July 2012,” sold for $760,000 on $300,288 of annual profit and roughly $374,000 of revenue, brokered by Stephanie Veal to a US buyer in January 2026. The business with 2.84× the earnings fetched 55.92% of the price.

Flippa does not hide the reason; it prints it. The kitchenware card says a business ten months old “has generated almost no track record of sustainable earnings,” and that “buyers priced the asset accordingly, applying a steep discount to reflect execution risk.” The blog card says the opposite in one clause — fourteen years of operation, 2.53×. Read the two together and the line item being transacted is not the cash flow. It is the date the thing started.

One pair proves nothing on its own, so take the marketplace’s own aggregate. Flippa’s Digital M&A Insights Report, H1 2026, written by Tony Xu, its Head of Product, covers 1 January to 30 June 2026 and prints average profit multiples by model: marketplace 3.33×, app 2.62×, SaaS 2.47×, content 2.32×, media and community 1.63×, YouTube 1.57×, ecommerce 1.55×, service and agency 1.43×. The whole distance from the cheapest model to the dearest is 2.33×. The distance between the two kitchen deals, inside what a buyer would call one category, was 5.07×.

That is the finding, and the same report states it as a structural fact rather than an accident: top-quartile assets sold at “1.6× to 2.7× above the category average.” The variance inside a category is larger than the variance between categories. Choosing what to build moves a seller across a 2.33× range. Where they land in their own category’s distribution moves them across a wider one, and nothing in the popular advice is addressed to the second number.

Now the part that inverts the usual reading of size. Flippa’s table of multiples by transaction size: $10K–$100K average 2.24×, top quartile 5.96×; $100K–$250K 1.85× and 3.82×; $250K–$1M 1.82× and 2.84×; $1M+ 2.50× and 5.42×. The smallest band on the board carries the highest top-quartile multiple of any band — above the million-dollar band — and the widest ratio of top quartile to average, 2.66× against 2.17×, 2.06× and 1.56×. The received wisdom that multiples expand with size is true of the averages and false of the tail, and the tail is the only part a seller can aim at.

Which matters here because the small band is the only one a person with $500 can be in. Flippa’s own sell page confines its self-service packages to businesses with an asking price “up to $100K,” prices them from “US$29 for a 60-day listing,” and charges “10% of the final transaction value” only when a sale completes. Acquire.com, the other venue, charges sellers 8% below $250k plus $25 a month to list. The entry fee to stand in the highest-dispersion band in digital M&A is twenty-nine dollars.

So the trade has two sides and both are posted. Sell into the top quartile of the $10K–$100K band at 5.96×; buy with the proceeds at the cheapest average on the model table, service and agency at 1.43×. Net of Flippa’s 10%, one dollar of annual profit sold buys 3.75 dollars of annual profit bought. And the buy side is not theoretical: broker Marco Reeves wrote up an email and SMS agency exit on 1 July 2026 with $382,543 of trailing revenue, $262,551 of net profit, a 69–75% margin and client acquisition cost “effectively $0,” which closed at 1.6× against 426 interested buyers and three final bidders.

Here is the reading nobody holds. Every hour of effort in this market goes into the numerator — more revenue, more profit, a better product — because that is the part a founder controls. Flippa’s own closed-deal cards say the denominator is doing more of the work: $851,412 of real, banked annual profit was discounted to $425,000 for being ten months old, while $300,288 was marked up to $760,000 for being fourteen years old. A person with $500 cannot buy revenue. They can buy a start date, for $29, and a start date is what this market is actually pricing.