Martins Three Laws
Thirty years of strategy work, distilled by Martin into three laws. They function as a rapid audit, and most documents called strategies fail at least two.
Law I: Integration beats any single choice. No individual answer wins; the reinforcing system of answers wins. A brilliant Where to Play with a generic How to Win loses to a modest pair that need each other. The whole Strategy Choice Cascade exists to force this property, because nothing in ordinary planning does.
Law II: the customer is the only judge. Strategy is scored by compelling customer action, not by internal elegance, board approval or analyst applause. If the customer does not move, the strategy failed, whatever the deck said. This is the law that disqualifies the most strategy work, because internal measures are so much easier to hit.
Law III: the opposite must not be stupid. Formalised as the Opposite Test: if no sane rival would choose the reverse of your "choice", you have not made one, you have stated an operating imperative shared by everyone.
Using the laws as an audit
Take any strategy document and ask the three questions in order. Is it integrated, do the choices strengthen each other or merely coexist? Does it move a customer, is there a named behaviour change the strategy predicts? Is it actually a choice, would the Opposite Test pass? The specimens that fail are catalogued in Strategy Impostors, and the posture that fails all three at once is Playing to Play.
How this connects to OKRs
The laws translate directly into goal quality. Law I is the test of an OKR portfolio: do the quarter's bets reinforce each other and the cascade, or merely coexist? Law II is the deepest argument against task-based Key Results: initiative completion is internal theatre unless customer behaviour changes. Law III belongs in every planning session: run it on each draft Objective before anyone falls in love with the wording.
One line to keep: Run the three laws on every draft Objective: is it integrated, does it move a customer, is the opposite stupid.
Our synthesis of Roger Martin’s published work, sources credited. Read the originals: they’re excellent.
How do I use Martin's three laws on OKRs? +
As a three-question audit on every draft Objective: is the portfolio integrated so the bets reinforce each other, does it move a named customer behaviour, and would the opposite be stupid? Most weak OKRs fail at least one.
How is this different from the Opposite Test alone? +
The Opposite Test is Law III on its own. The three laws add integration (Law I) and customer action (Law II), so together they catch coexisting goals and internal-milestone Key Results that the Opposite Test would happily pass.
Reading about method is not the same as running it. We install this system and build the capability that stays.
Turn strategy into executed strategy →