Operations.

The framework

1. 01

   What is this operation actually optimised for?

   The mission is one sentence naming what the operation will be best at and, as a consequence, what it will only be adequate at: we deliver standard work faster than anyone, and we don't take bespoke; we do bespoke superbly, and we're never the cheapest. The second half is the part that gets deleted, and it's the part doing the work. Day 79 made this case for a market position, the refusals that keep a strategy defensible. Skinner's version is narrower and more physical — it's about what a production system can be good at, whatever your positioning says. Write the sentence, then read the second half aloud to whoever sells for you.

2. 02

   Cheap, fast, bespoke, flexible — and what each one takes from the others.

   These aren't four dials you can turn up together, and the conflicts are mechanical rather than attitudinal. Buying speed means holding stock or holding capacity idle, and both land as cost. Taking bespoke work means short runs and improvisation, which is the opposite of the repetition that makes anything cheap. Flexibility means slack, and slack is the first thing a cost target removes. So rank the four for your own operation, first to last, then check that ranking against where your money and your best people actually go. In most businesses the written ranking and the spending ranking disagree, and the spending one is the real mission.

3. 03

   Split the place before you split the strategy.

   Skinner's plant within a plant is the practical move when two missions genuinely both have to exist. Give each its own space, its own people, its own scheduling and its own measures, even inside one building and one P&L. In a services firm that means the retainer team and the project team don't share a queue; in a kitchen it means delivery orders don't compete with the dining room for the same pass. Separation costs a little efficiency on paper and buys back far more in the interruptions it stops generating.

4. 04

   One set of measures across two missions punishes both.

   The clearest sign of an unfocused operation is one scorecard applied to incompatible work — the bespoke team judged on utilisation, the standard line judged on responsiveness. People game whichever number is watched, and the mission that fits the metric worst gets quietly starved. Give each mission its own two or three measures and review them apart. The trade-offs aren't absolute — Kasra Ferdows and Arnoud De Meyer argued in 1990 that quality, dependability, speed and cost build on one another in that order — but cumulative gains run out, and past that point you're still choosing.