Reversible in Principle, Unrecoverable in Practice
A decision can be technically undoable and still be governed, in effect, as though it were permanent — because the cost of undoing it is organizational, not technical.
Reversibility is usually assessed as a binary technical property: can this action be undone. Governance treats reversible decisions as lower-stakes — worth making quickly, without exhaustive information gathering — because reversal is available as a correction mechanism if the decision proves wrong. This is a reasonable heuristic. It is also incomplete in a way that matters.
The distinction. Technical reversibility — the mechanical ability to undo an action — is not the same as organizational reversibility — the practical willingness and capacity to actually undo it once reputational, relational, or structural consequences of the initial action have begun to accumulate. A contract can be technically terminable. A public commitment can be technically retractable. A staffing decision can be technically reversed. In each case, “technically” is doing significant work, because the act of reversal itself produces consequences — reputational cost, relationship damage, internal disruption, market signal — that are often larger and harder to absorb than the consequences of simply continuing.
The mechanism. Reversal is not necessarily costless. It can require the organization to publicly or internally acknowledge that the original decision was wrong, creating costs independent of whatever the decision was about. Trust, once damaged by an initial decision and its early consequences, is difficult to fully rebuild even after formal reversal. This is why organizations can continue down a path even after reversal has become more attractive in principle — not because reversal is unavailable, but because the cost of reversing has grown to exceed, or appear to exceed, the cost of continuing.
Governance consequence. A board that approves a decision on the reasoning that “we can always undo this if it doesn’t work” is relying on an assumption that may quietly stop being true the moment the decision is executed and its early consequences begin to accrue. The reversibility that justified the speed of the original decision may no longer exist in practice by the time reversal is actually being considered.
What this changes. The governance question is not simply “is this reversible” but “what would reversal actually cost once this decision has been in effect for some period, and is that cost being honestly priced into the original approval, or is reversibility being treated as a permanent, static option available at no cost whenever needed.”
Evidence boundary. The general proposition — that reputational and relational consequences can make formal reversal practically costly even where it remains technically available — is well supported across research on organizational trust, reputational risk, and crisis recovery. The specific governance framework proposed here, for pricing reversal cost into initial approval, is Obraval’s interpretation and synthesis rather than an established practice.
The executive question. If we had to reverse this decision six months from now, what would that reversal cost us that has nothing to do with the decision itself?