The Broken Equation Test
No amount of product craft can fix a company whose underlying math doesn't close.
- Difficulty
- Advanced
- Time to result
- ~weeks to results
- Steps
- 5
- Confidence
- 95%
Tom Conrad (Quibi's CPO, Pandora's long-time CTO, engineering leader at pets.com) spent a career believing that if you solve a real problem ten times better than the alternatives, word of mouth handles the rest. Quibi taught him a second, senior truth: a company is also a math problem — capital in, returns out, over some horizon. Product decisions are variables inside that equation, but if the equation itself is broken, iteration and execution cannot rescue the outputs. The framework is to stress-test the equation before you commit to the execution.
Origin
Tom Conrad's post-mortem on Quibi, cross-referenced with his pets.com experience. Quibi bet ~$2B could buy enough bespoke premium content to acquire and retain subscribers; Conrad concluded the true number was six to ten billion, which no one could stomach.
Core principles
- 01Great product execution is a variable in the equation, not the equation.
- 02An excess of investment can be its own albatross — it licenses irrational decisions.
- 03Ask 'how much would it actually take?' before asking 'how well can we build it?'
- 04The equation must survive a shock — a format that only works with the perfect production pipeline is a fragile equation.
- 05Timing is a term in the equation: the same idea can be broken in 1999 and worth $9B in 2019.
- 06Winding down early and returning capital can be the correct move; spending the last penny on a fruitless salvage is not.
How to run it
- 1
Write the equation before the roadmap
State plainly: this much capital, poured in over this horizon, produces this much return — because of these specific assumptions. For Quibi: can $2B of bespoke content acquire and retain enough subscribers? Make the number explicit.
Pro tip Force the question into a single sentence containing a dollar figure and a time horizon. Vagueness in either is where broken equations hide.
- 2
Sanity-check the magnitude, not just the sign
Ask whether the required investment is off by an order of magnitude, not whether it's slightly optimistic. Quibi's real number wasn't $2.2B, it was $6-10B. A 4x error in the capital requirement is fatal in a way a 10% error is not.
Pro tip Compare against the best-executed comparable. Chewy grew brick by brick over a decade to $9B; pets.com tried to compress that with TV spend in 19 months.
Watch out Beware conclusions like 'this is just a stupid business'. Shipping dog food around absolutely can work — it just couldn't work when 80% of the country was on dialup. The idea wasn't wrong, the equation's timing term was.
- 3
Check whether capital abundance is distorting behavior
Three overfunded pet e-commerce sites each raised ~$50M, all assumed a zero-sum game, and all escalated into national broadcast advertising. The money itself produced the unwinnable arms race. Ask what decisions you are making only because you have the cash.
Pro tip Run the plan you'd run with a quarter of the money. If it's a better plan, the money is making you worse.
Watch out 'We must, because they are' is the signature of an unwinnable arms race.
- 4
Stress-test the fragile term
Identify the assumption whose failure collapses the whole equation and ask what happens if it breaks. Quibi's daily-content strategy — roughly a third to half of content spend — depended on purpose-built professional studios, produced day-of. Two weeks after launch, COVID meant it was all shot in hosts' garages, i.e. exactly like the YouTube content it was supposed to be differentiated from.
Pro tip The fragile term is usually the one that requires physical infrastructure or a partner you don't control.
- 5
Reserve the right to stop and return the capital
If market conditions make further funding impossible and the business needs it to reach profitability, wind down early and return the remaining balance rather than burning it on a salvage attempt. pets.com did not go bankrupt — it shut down and returned the balance to investors, which no public company had done before.
Pro tip Decide the wind-down trigger while you're still solvent and unemotional.
Watch out This is a leadership-team decision that must be made before the runway is gone — you cannot return money you've already spent proving a point.
In the wild
Quibi bet a couple of billion dollars would buy enough bespoke, phone-native premium content to get people to subscribe and retain. The team executed impressively — 70 shows in 18 months, more than all major broadcast networks combined make in a year. But the library required to sustain subscriptions was far larger, and the roadmap needed time to iterate on the content format itself. The math wanted $6-10B.
→ The product execution was real; the equation was broken. Betting $10B on an unproven format was more than anyone could stomach, and Quibi shut down.
Three pet e-commerce sites each raised an excessive ~$50M, all believed it was zero-sum, and all escalated into irrational national broadcast television spend. The company went from nothing to public to out of business in about 19 months — but the leadership team shut it down and returned the remaining cash to investors rather than going bankrupt.
→ Chewy, executing brick by brick over a decade in a later era, was acquired by PetSmart for $3B and is worth ~$9B today. Same category, different equation.
Common mistakes
Assuming product excellence rescues bad math
The belief that solving a problem 10x better means word of mouth handles the rest is true only when the equation closes. If the required capital is 4x what you raised, no amount of delightful iteration changes the failed output.
Reading excess funding as an advantage
An excess of investment can be its own albatross. It funds irrational escalations (broadcast TV arms races) and lets you skip the discipline of proving the math on small money first.
Confusing the idea with the equation
Critics said pets.com was a stupid business — you can't ship dog food. That's wrong; you can. The failure was in timing and capital structure, not the concept. Killing good ideas for the wrong reason is its own expensive mistake.
Is it for you?
Best for
Founders, CPOs, and investors evaluating a capital-intensive big swing — a new category, a content library, a marketplace requiring physical operations — before committing to build.
Not ideal for
Small, cheap, reversible product bets where the cost of just trying it is lower than the cost of modeling it.
From the transcript
“companies are also kind of like a they're kind of a math problem”
“no amount of like iteration and execution can like kind of get you out of the the failed outputs of the broken equation”
“happened is pretty quickly it became clear that the math was just wrong it wasn't going to take two billion it was going to take…”
“having an excess of of investment can be its own Albatross”
From the episode
Failure