Every valuation argument eventually reaches for a list of comparable sales. The list is usually long, impressive and largely unusable as it stands.
Public sales data is a by-product of people choosing to report their deals. That makes it useful, but it also gives the data a shape: it over-represents some kinds of sale and almost entirely misses others.
What to take out first
- Duplicates. The same sale reported by the buyer, the seller and the marketplace often appears three times at three slightly different prices.
- Related-party deals. A transfer between two companies with the same owner is not a market price.
- Bundles. A name sold together with a website, a business or other names says little about the name alone.
- The wrong extension. A sale in another extension is evidence about the word, not about your name.
What is missing
The largest sales are often the ones that are never reported. Companies buying names for a launch or a rebrand tend to insist on confidentiality, and those are exactly the deals that set the top of the market.
So a clean list of public comparables is a floor, not a ceiling. It tells you what has been paid in the open. It does not tell you what an operating company with a deadline has paid in private.
The comparable you can see is rarely the one that sets the price.
Dating the evidence
Old sales need adjusting before they can be compared with today. Demand for particular words moves with the industries that use them, and a sale from a year when a sector was raising money freely says little about the same sector after the money has gone.
In PrismIndex, every comparable carries its date, extension and a reliability weight alongside its price. When a valuation leans on one sale more than the others, the report says which one, and why.
