Triangulate the Customer's Truth
Don't take what customers say literally — reconstruct their economics and incentives independently, then find the urgent one
- Difficulty
- Moderate
- Time to result
- ~weeks to results
- Steps
- 4
- Confidence
- 90%
Droege's approach to customer discovery: never trust the customer's stated numbers or stated problems literally. Instead, build your own independent 'ground truth' of their economics, read their real (often non-financial) incentives, and prioritize the one problem urgent enough to top their daily thinking.
Origin
Developed by Jason Droege while launching Uber Eats, when restaurants wouldn't share their unit economics and he had to reconstruct them from scratch.
Core principles
- 01Customers' stated incentives are not always financial — ego, career growth, and trust also drive them
- 02Show me the incentive and I'll show you the outcome
- 03Independent ground truth beats self-reported numbers
- 04Value that isn't the customer's most urgent daily concern won't get bought
- 05'Do what they mean, not what they say' — plus, read the underlying incentives
How to run it
- 1
Distrust the stated answer
Treat every customer statement — numbers, problems, priorities — as a claim to verify, not a fact. Constantly question what you're hearing at the start.
Pro tip Assume suspicion runs both ways: they may withhold real numbers because they don't trust your motive.
- 2
Reconstruct ground truth independently
Build your own model of the customer's economics from first principles rather than relying on what they'll tell you. Buy the product, decompose it, and compute the numbers yourself.
Pro tip Triangulate three sources: your independent reconstruction, what the customer tells you, and third-party/industry data. Overlap = a real insight.
Watch out Insiders will resent the model because reality is messier — use it for confidence, not as gospel.
- 3
Read the non-financial incentives
Map what the buyer actually needs to say yes — often ego, career safety, or trust in you, not just ROI. Identify what they need to hear and what you must supply to unlock the deal.
- 4
Solve the URGENT problem, not the biggest one
Among real problems, target the one that tops the buyer's busy day. A big annual problem that isn't top-of-mind daily leads to 'a long road to a small town.'
Pro tip Ask what they check every morning — did people show up, did I make money yesterday — that reveals the urgent lever.
Watch out You can build something genuinely valuable and still fail if it isn't the customer's most urgent concern.
In the wild
Restaurants wouldn't share unit economics and were suspicious of 'these Uber guys' asking about ham costs. So Droege's team ordered food, got a supplier base catalog, and physically matched how much the ham, cheese, bread, and lettuce weighed to compute ingredient vs. labor cost themselves, then triangulated against what restaurants and industry sources said.
→ They derived that restaurants pay ~20-30% to ingredients, ~20-30% to labor, ~10% to real estate, giving them confidence to price the take rate near 25-30% and structure the whole marketplace.
Common mistakes
Building for the stated big problem, not the urgent one
A restaurant will say ingredient cost (20-30%) matters most, so you build a tool saving 10% on ingredients — but that's an annual concern, not their daily urgency, so it never gets prioritized or bought.
Taking customer numbers literally
Customers give evasive or self-serving figures; without an independent reconstruction you can't tell what's real, and you'll price or design against fiction.
Is it for you?
Best for
Founders and PMs entering an unfamiliar industry where customers won't share honest economics
Not ideal for
Well-understood markets with transparent, reliable public unit economics where reconstruction is wasted effort
From the transcript
“I don't take what a customer says like literally”
“I look at the underlying incentives of the customer and the underlying incentives of customers are not always financial. Sometimes it's ego, sometimes it's career…”
“we just matched up like how much does the ham weigh, how much does the cheese weigh, how much does the bread weigh, how many…”
“the biggest thing people miss when they're building new products is the urgency of the buyer part of it”
From the episode
First interview with Scale AI’s CEO: $14B Meta deal, what’s working in enterprise AI, and what frontier labs are building next
Jason Droege