── ── Cognitive bias
Sunk-Cost Fallacy
The Sunk-Cost Fallacy is the tendency to continue an endeavor because of resources already invested, even when continuing no longer makes sense. Money, time, and effort already spent are gone regardless of what you choose next. A rational decision weighs only the marginal cost and benefit from this point forward.
How it works
Past investments are sunk — they cannot be recovered by any future choice, so they should carry zero weight in deciding what to do next. The only valid question is whether the remaining cost is worth the expected remaining benefit, judged from today.
The fallacy persists because abandoning an investment feels like admitting waste and loss, and humans are powerfully loss-averse. The fix is to mentally separate the unrecoverable past from the still-open future and decide only on the latter.
When to use it
- Deciding whether to keep building a feature or product that isn't gaining traction
- Reconsidering a costly project, hire, or partnership that's underperforming
- Evaluating whether to keep pouring spend into a stalled marketing channel
- Facing a pivot decision after significant time and capital are already committed
When not to use it
Not a reason to abandon things at the first setback — distinguish a genuinely poor forward outlook from normal early difficulty that perseverance would overcome.
Worked example
The Concorde and the cost of not quitting
Britain and France continued funding the supersonic Concorde airliner long after it was clear the project would never be commercially viable, in large part because so much had already been invested and abandoning it felt like wasting that money. The pattern is so emblematic that the sunk-cost fallacy is sometimes called the 'Concorde fallacy.' The prior spending was gone either way; only the forward economics should have mattered.
Why it matters for founders
Few decisions are harder than killing a product you've poured a year into, and the sunk-cost fallacy is exactly what keeps doomed projects alive past the point of reason. The discipline is to judge every commitment on its forward economics, as if you were deciding to start it today. deciqAI's agents evaluate decisions on marginal cost and benefit from here forward, so prior investment informs the lesson without dictating the next move.
Install this skill (free, MIT)
npx skills add deciqAI/knowledge-skillsFAQ
How do I tell a sunk-cost trap from healthy persistence?
Ask whether you'd start this today knowing only what you know now and ignoring what's already spent. If the forward case is strong, it's persistence; if you're continuing mainly to justify past investment, it's the fallacy.
How is this different from opportunity cost?
Sunk cost concerns past, unrecoverable spending that should be ignored; opportunity cost concerns the future alternatives you forgo by continuing. Both point you to evaluate decisions on forward economics only.