The reason a plan survives contact with nobody

The roadmap was presented to the whole company in March. By June three of its assumptions are known to be wrong, and the roadmap is unchanged. Nobody is defending it dishonestly. It has become a commitment that exists independently of the conditions it was built for.

What everyone sees

Sticking to a plan is read as discipline, and abandoning one as flightiness, so persistence is rewarded by default. That accounting has no term for information arriving after the announcement. It treats the plan as the standard against which reality is judged, rather than the reverse.

What is actually happening

Buehler, Griffin and Ross documented the planning fallacy, finding that people systematically underestimate completion times even when they know their own history of doing so, because prediction is built from a constructed scenario rather than from past distributions. Gollwitzer, Sheeran, Michalski and Seifert added a second mechanism specific to announcement: participants who told others about their intentions subsequently acted on them less than those who kept them private, apparently because public declaration already delivered some of the sense of having become the sort of person who does that thing. Together they describe a plan that is optimistic when written and partly discharged by being read out. Its survival afterwards owes little to its accuracy.

Why it stays hidden

The mechanism hides because revision is expensive in a way that persistence is not. Changing the plan requires an explanation, a meeting and an implicit admission; leaving it in place requires nothing and produces no document. So the asymmetry in effort is mistaken for an asymmetry in evidence, and the plan outlives its assumptions by default rather than by argument.