Before the method
Why do any of this?
Nobody bets on a market believing it will not work. Every failure below was somebody's confident decision — so confidence is not what was missing.
of failed startups die from “no market need” — the single most common reason.
CB Insights
moves into a new market or segment actually succeed.
Bain & Company
of market pioneers ultimately fail — being first is no guarantee.
Tellis & Golder
Moving into a new market, or going after more share of one you are already in: either way, somebody in the room has a number, a view on what customers want, and a reason to believe you will win. The question is whether any of it is written down where a colleague could check it.
“We're small. This is overkill.”
Analysis is expensive. So is guessing — it just bills later, as six months building the wrong thing. Only one of the two you can see coming and cap. And only one pays back: every move after it is made on what it found.
“We already have a process.”
You have a process. Could you reconstruct how any one analysis was done — how that market got sized, how those needs were verified? It is real work, and it is in somebody's spreadsheet and somebody else's deck.
Lovallo & Kahneman, HBR, 2003
“We listen to our customers.”
Ask what goes wrong and they are the best source there is. Ask what to build and they describe what they already have — or something they would like and would not pay for.
Schmidt & Bijmolt, JAMS, 2020 — stated willingness to pay runs ~21% above real
“We know how big this market is.”
You know one number, from one direction, for today. You arrive two or three years out, into whatever the market is by then — so build it twice, and say what is driving the growth.
“We use AI for this now.”
So do we — it's your partner here. The difference is what you are left holding: a figure with its source attached, built the same way each time, not five people's chats nobody else can re-run — and a method underneath it that is established strategy practice, not whatever the model reached for.
This is not only an argument. Firms that do it well do better, across tens of thousands of them.
Companies that systematically know what their customers value out-perform those that do not.
214 correlations, 36,150 firms · r = .32 · Kirca, Jayachandran & Bearden, Journal of Marketing, 2005
Planning is associated with better performance — in small firms as well as large ones.
183 samples, 30,246 organizations · r = .20 · Hamann, Halw & Guenther, Strategic Management Journal, 2023
A product with a real, differentiated customer benefit is the strongest single predictor of commercial success — and the up-front homework that finds one is what separates winners from losers.
203 launches, successes and failures alike · Cooper & Kleinschmidt, J. Product Innovation Management, 1987
And the tools it gets compared to
Nothing else is close.
Two questions decide it: how much of the answer is actually yours, and what you still have when the project ends. Every tool people reach for was built for a different job — the nearest one answers one rating of nine. Under each one, why it sits where it does.
The smallest honest version is an afternoon.
- Name the outcomes customers want — not the things you would sell them.
- Score each one: how much it matters, how well it goes today.
- Size the market twice, from different directions. Say how fast it is growing.
- Write down what would change your mind, before you go looking.
The last one is what makes it an analysis instead of a case for what you already decided.
A market is rarely just good or bad. It is good for somebody, at some price, through some way in — and the work is what tells you whether that somebody is you. Sometimes it tells you no, and learning that in an afternoon is the cheapest thing here. Either way you finish holding needs sharp enough to argue with, a size defended from two directions, and your assumptions out where somebody can check them — which is what the next move gets argued from, instead of starting over.