Inversion
5 min read
Core idea
Approach the problem from the opposite end
To invert is to turn a thing upside down. As a thinking tool, inversion means starting at the end instead of the beginning, or proving the negative instead of the positive. The nineteenth-century mathematician Carl Jacobi summarized it in three words: invert, always invert. When the forward path is blocked, foggy, or crowded with competitors, the inverse path is frequently open. Most problems become tractable the moment you let yourself work backward from "what would the world look like if this were true?" instead of forward from "how do I make this true?"
Avoiding failure beats chasing brilliance
Charlie Munger's lifelong rule was that he tried not to be brilliant; he tried to avoid the standard ways of failing. The two are not the same. Chasing brilliance requires you to be right about something hard; avoiding failure only requires you to recognize a small catalogue of known traps. Most domains — investing, marriage, careers, surgery — have a long list of well-documented ways to lose and a much shorter list of ways to win. Inversion turns your attention to the long list. Subtract the failure modes and the residue is often surprisingly close to success.
Why it matters
Forward thinking shares space with everyone else's forward thinking
When everyone in a market, hiring pool, or scientific field is reasoning forward, the marginal value of one more forward thinker is small. The competitive advantage sits with the person who runs the same problem in reverse. John Bogle did not try to beat the market — he asked how investors lose money (fees, churn, poor manager selection) and built the index fund around minimizing those losses. The result was one of the most successful financial products in history, built entirely from inverted questions no one was bothering to ask.
Some problems only yield to the inverse
The Greek mathematician Hippasus could not derive the square root of two by forward division, so he flipped the problem — what would the number have to not be? — and discovered the first irrational number, overturning the Pythagorean worldview. Many proofs in mathematics, most safety engineering, almost all good ethics, and a good portion of strategy live on the inverse side: easier to show what cannot be true, what must not happen, what we must not do.
Key takeaways
Mental model
Practical application
There are two distinct ways to invert. Pick the one that fits the situation.
-
Inversion by assumption. State the conclusion you want and work backward. "If our product had a million paying users, what would have to be true about onboarding, pricing, distribution, retention, and word-of-mouth?" Each answer becomes a checkable sub-problem. This is Edward Bernays asking what would the world have to look like for women to smoke? before asking how to sell more cigarettes — and then setting about making that world real.
-
Inversion by failure mode. List every well-known way the thing could fail and decide what would prevent each. For a marriage: distance, contempt, financial secrecy, neglect. For an investment portfolio: high fees, leverage, concentration, churn. For a software launch: silent data loss, opaque errors, broken auth on day one. Solve the list and the residual is usually success.
-
Make the inverted list before the forward list. The order matters. If you brainstorm "ways to succeed" first, you anchor on aspirations; the failure list afterward becomes a polite footnote. Do failure modes first and the forward plan tightens around the real risks.
-
Re-invert when the world changes. A failure mode that does not exist today (remote-team coordination collapse, prompt-injection attack, supply-chain seizure) becomes a real one next year. Periodically re-run the inverted question on your current plans.
-
Pair inversion with the pre-mortem. Before committing to a major decision, sit the team down and pretend it is two years out and the project has failed catastrophically. Write the failure story. The list of plausible causes the team generates is almost always the list you need to address.
Example
Designing a remote-work policy by what must not happen
A 200-person company is writing its first remote-work policy. The forward question is "how do we make remote work great?" — a question vague enough to produce a vague document. The leadership team tries an inverted question instead: what would make remote work fail catastrophically here in eighteen months?
The list is concrete. Coordinating across five time zones with no overlap. Loss of informal mentorship for junior staff. Performance reviews that quietly favor whoever happens to be in the room. New hires who never meet their team in person and quit at six months. Decisions made in Slack threads no one can find later. Burnout from blurred work-home boundaries. Two-tier culture between an "HQ" and "satellites."
Each failure mode points to a concrete policy: a four-hour daily overlap window required for every team; an explicit mentor assigned to every hire under two years' tenure; a structured promotion rubric not tied to physical visibility; a mandatory once-per-quarter on-site for distributed teams; decisions written into an indexed decision log within twenty-four hours; explicit no-meeting blocks and end-of-day signals; one common video-first meeting norm with no satellite vocabulary.
The resulting document is shorter and sharper than any forward draft would have been. It does not promise that remote work will be wonderful. It removes the seven specific ways it would have quietly broken — and that is enough to make the rest work.
Related lessons
Related concepts
- Inversionlinked concept
- Mental Modelslinked concept
- Avoiding Failurelinked concept
- Via Negativalinked concept