The hidden cost of pilot-itis in public-sector programs
Endless piloting is rarely caution. More often it's an institutional refusal to learn.
A program that has been “in pilot” for six years is no longer a pilot. It is a permanent state of organisational ambivalence.
The cost of pilot-itis is not the money spent on the pilots. It is the avoidance of the decision the pilots are supposed to inform. Each successive pilot defers the question — “does this work at scale?” — without ever answering it.
Why it persists
Pilots are politically safe. They commit nothing. They acknowledge a problem without proposing a solution. They generate activity that can be reported as progress. They protect the leadership from being responsible for a real decision.
A pilot in year one is prudent. A pilot in year five is theatre.
The cost
Three programs we have evaluated had been “in pilot” for between four and seven years. Two of them had strong year-one evidence that they worked. Both could have been scaled in year two if the leadership had been willing to commit. Both instead ran annual pilot variations that produced steadily diminishing learning.
The opportunity cost is the program at scale that was never built. The financial cost is the recurring evaluation budget. The cultural cost is harder to measure: an organisation that learns it can defer decisions indefinitely loses its ability to make decisions at all.
What to do
Set a sunset date on every pilot at the start. Year three is “scale or kill.” The decision rule is named at design time, not negotiated after results come in. Evidence either crosses the threshold or it does not. Either outcome is a useful conclusion.
Article body — approximately 900 words.