Money, time or effort already spent and unrecoverable regardless of what happens next. Rationally irrelevant to any forward-looking decision, and rhetorically decisive in a remarkable number of them.
No committee ever votes to "throw good money after bad". They vote to "protect the investment already made", which is the same decision wearing better clothes.
Last reviewed 3 July 2026 · Free and ungated
Operators with no stake in the history can tell you whether the forward case stands on its own. Confidential, and a complimentary first report.
Get independent perspectivesA sunk cost is expenditure that cannot be recovered whatever you decide next: the two years and eight million already spent on a platform build are the same whether you continue or stop. Decision theory is unambiguous that only future costs and future benefits should bear on the choice. Almost no organisation actually behaves this way.
The phrase persists because it is not really an economic argument; it is an accountability argument. Stopping converts a quiet, distributed loss into a visible one with a date and an owner, and the people who approved the original spend are usually the people in the room deciding whether to continue. Continuing defers the reckoning and keeps every reputation intact for another quarter. Understood that way, sunk-cost reasoning is less a thinking error than an incentive problem, which is why appeals to rationality rarely fix it and why the question "would we start this today, knowing what we now know?" so often has to come from outside the room.