The checklist was never the intervention.

Copy the artefact. Audit the compliance. Wonder why nothing moved.

Almost everything a design team produces is an artefact: a screen, a form, a template, a checklist. It is natural to assume the artefact is the thing that works, because it is the thing you can hold. The most famous checklist in medicine says otherwise.

The same piece of paper, twice

The paper was the occasion, not the intervention

What worked in the first eight hospitals was the briefing, the habit of checking, and the fact that a junior nurse now had explicit permission to stop a senior surgeon.

Ontario made the paper compulsory and audited the signatures. It got signed pieces of paper. The behaviour that produced the original result was never copied, because nobody had named it as the thing that worked.

This isn’t confined to hospitals. Schools get it with mandated lesson-plan templates, when what made the pilot work was the planning conversation between two teachers. Product teams get it with design systems, when what made the first team fast was a shared habit of reviewing work together.

Before you copy something that worked, name the behaviour it was the occasion for.

The best evidence sits on the least valuable work

Go back to the gas fitting from the last piece. Nobody has run a trial on it, and nobody ever will. You don’t need evidence for something that can’t happen.

That creates a real problem. The cheap things to test are the low-consequence things. So the strength of the evidence runs roughly opposite to the size of the effect.

• Judge design by asking “where’s the study?” and you will reliably fund the wrong work.
• Argue for design using studies and you will reliably argue for the wrong work.

Consequence and credit vary independently

Two things decide what a design decision is worth.

• Consequence: does the decision sit on the path by which the system actually fails, and does it repeat?
• Credit: can anyone see the effect, attribute it and count it before the next performance review?

Consequence → Credit: can anyone see and count it? →

The value is hereDefaults, interlocks, standards, purchasing specifications. A failure that never happens is recorded nowhere.

Rare and preciousOutcomes already measured for other reasons, like infection rates. Improvements land on a chart within months.

QuietSmall things nobody notices, and nobody needs to.

The evidence is hereA visual refresh that lifts sales half a percent. A clean chart within a fortnight.

HighLowInvisibleVisible

So the profession has arranged itself around the work it can count, then objects when executives treat design as the function that produces small numbers. The executives are reading the evidence the profession handed them.

Nobody is promoted for a disaster that failed to occur

Los Angeles wrote a schools technology specification that a federal review later found had effectively excluded every supplier except Apple. It then spent about $1.3 billion acting on it. The iPads came preloaded with a curriculum that 2 of 69 trial schools ended up using.

The specification was written before anyone with a design title was involved, and it fixed the outcome more completely than any later decision could. One person rewriting that document could have changed the whole programme. Nobody would have been able to see they had done anything.

Purchasing documents, standards, defaults and interlocks are the highest-leverage objects in most large organisations. Deciding once, for everybody, is what a headquarters is for.