Four Thousand Customers, and None of Them TwiceNarrow moat

The concentration question, asked of Dom Development, has a one-word answer: none. No buyer reaches a tenth of a percent of revenue, let alone ten percent. The largest customer in any year is a family purchasing a single flat — 4 448 of them in 2025, a record that included 1 232 in the fourth quarter alone¹.

That is a real form of safety and worth stating plainly. Dom cannot lose a quarter of its revenue because one counterparty changed its plans, cannot be squeezed on terms by a buyer with purchasing power, and never has to disclose that a related party has become its biggest source of sales. Several companies in this collection would trade for that position.

What it costs is everything on the other side of the ledger. A software company sells once and collects for a decade. Dom sells once and collects once. There is no renewal to protect, no expansion revenue, no switching cost — a customer who has just bought a Dom flat is, by definition, out of the market for one. The repeat business the company does get comes indirectly, through reputation and recommendation, and the moat's Quality and Repeat Buyers page deals with that.

The practical consequence is that demand has to be re-created continuously rather than retained, and the cost of doing so is a permanent operating expense rather than a one-time acquisition cost. Dom's answer is to make that expense unusually low, by being the name a Warsaw buyer thinks of first and by having flats ready when others do not.

The number that tests this is the marketing and selling cost per unit sold. If a company genuinely has to buy its customers afresh every year, that figure is its true customer-acquisition cost — and if it starts climbing while volumes are flat, the brand is doing less work than the essays claim.