A market size can be arithmetically perfect and still nonsense — a SOM bigger than its SAM, a bottom-up number that implies customers paying more than the value they receive. These are the errors that survive a spreadsheet review and die in a board room. So before any figure ships, it clears five automated gates. Here they are, each with the exact rule and a worked pass, run on the warehouse-robotics sample.
The funnel must strictly shrink: SOM ≤ SAM ≤ TAM. A SOM that exceeds its SAM means the narrowing logic is broken — you're claiming to obtain more than you can serve.
A sector cannot charge for more value than it creates. The bottom-up TAM must sit inside the addressable value pool — otherwise the price or attach-rate assumptions are impossible.
If this fails, the fix is almost always the bottom-up price or attach rate, not the pool. Detail in the value-theory guide.
The top-down capture share must be a minority of the parent market — at most ~60%. Above that, you've stopped sizing a sub-sector and started claiming it is the parent category, which is a framing error, not a market size.
The price-per-adopter implied by the top-down number should be within an order of magnitude of the bottom-up price assumption. If the two big methods disagree on price by more than 10×, one of them has a broken input.
This is the check that most often catches a silently-inflated top-down number — see the top-down vs bottom-up guide.
A three-year SOM should be a modest share of its SAM. Capturing more than ~30% of a served market in three years implies near-monopoly, and investors discount it on sight.
The instant estimate applies the same monotonicity and SOM-share logic live and flags the agreement band as you change inputs.
Free instant estimate → Commission a cited report