CONCEPT
The theory and practice of designing a working life.
A career strategy is a stated bet about what compounds for you, paired with a decision about what you will deliberately be bad at. Richard Rumelt's structure applies without modification: a diagnosis of the situation, a guiding policy, and a coherent set of actions that follow from it. A list of goals is not a strategy. Neither is a ladder. A ladder is someone else's strategy that you have agreed to be an input to.
Careers are portfolios of assets with very different half-lives, and most people never sort theirs.
Compounding: judgment in a domain, which improves with reps and transfers across employers. Reputation, which grows within a network and resets outside it. Relationships, which decay slowly and cannot be acquired quickly under pressure. Ownership, which compounds hardest and which most people never acquire at all.
Decaying: proficiency in a specific tool, knowledge of one company's internals, being current on anything. These are worth holding because they pay now, but they have to be re-earned continuously, and a career made entirely of them is a treadmill with a salary attached.
The strategic question is not what you are good at, it is which of your assets keeps paying after you leave the building.
The useful framing comes from bandit problems. Early on, information about yourself is worth more than optimization, so variance is cheap and breadth is rational. Later, accumulated position raises the cost of switching, and concentration pays better than sampling.
Both phases are correct in their regime. The failure is not exploring too long or committing too early in the abstract. It is never noticing that the regime changed, and running an explore policy at forty because it worked at twenty-five, or an exploit policy at twenty-five because it looks like seriousness.
The only reliable detector is a review with a fixed cadence. Once a year, written: what compounded, what decayed, what I learned about my own capability, what I would now be unwilling to give up. Absent that, drift feels identical to strategy from the inside.
*Follow your passion* inverts the causality. Cal Newport's argument in *So Good They Can't Ignore You* is that rare and valuable skills come first and interest tends to follow capability, rather than the other way around. Passion is more often a result of getting good at something than a reliable input to choosing it.
*Make a five-year plan* offers precision over variables you do not control. Direction plus a review cadence survives contact with reality; a plan mostly generates the feeling of having decided.
*Maximize compensation* treats money and optionality as the same asset. They are not. A high salary that depends on one firm, one region, or one non-transferable body of knowledge is a purchase of lifestyle paid for with flexibility. That can be the right trade. It should be a trade you made on purpose.
The last error is the largest. Strategy is not a document; it is the sum of small allocations of the calendar. If last month's calendar does not show the bet, the bet was never placed.