Field Note: The Sunk Cost Trap
Date: 2025-01-15 Sector: DECISION-MAKING Read Time: 4 minutes
The money is gone. The time is gone. And yet, we keep spending more — because we've already spent so much.
The Observation
A company had spent 18 months and $400,000 building a custom CRM. It didn't work well. Users hated it. The vendor was unresponsive. But every time someone suggested switching to an off-the-shelf solution, the response was the same:
"We've already invested too much to walk away now."
So they invested more. Another 6 months. Another $150,000. The CRM still didn't work. Users still hated it. But now the argument was even stronger: "We've invested $550,000. We can't abandon it."
They eventually switched to Salesforce. Total wasted investment: $600,000 and 2 years. If they'd switched at month 6, it would have been $100,000 and 6 months.
The Trap
Sunk costs are costs that have already been incurred and cannot be recovered. They are irrelevant to future decisions. And yet, they dominate future decisions.
The rational question: "Given where we are now, what's the best use of our resources going forward?"
The emotional question: "But we've already put so much into this!"
The emotional question wins almost every time.
Why We Fall For It
Loss Aversion
Humans feel losses roughly twice as intensely as equivalent gains. Abandoning a $400,000 investment feels like losing $400,000 — even though the money is already gone whether you continue or not.
Identity
Projects become part of people's identity. The CRM project was "Maria's initiative." Killing it feels like a personal failure. So Maria fights for it, not because the data supports it, but because her reputation is attached.
Narrative
We need our decisions to make sense in a story. "We started this, invested heavily, and it paid off" is a good story. "We started this, invested heavily, abandoned it, and started over" feels like failure — even if the outcome is better.
Commitment Bias
Once publicly committed to a course of action, changing direction requires admitting the original decision was wrong. Most people and organizations would rather double down than admit an error.
How to Escape
The Clean Slate Test
Ask: "If I were joining this project today, knowing what I know now, would I invest in it?" If the answer is no, you should stop — regardless of what's already been spent.
Pre-Commitment to Kill Criteria
Before starting a project, define the conditions under which you'd kill it:
- "If we haven't hit milestone X by date Y, we stop"
- "If user satisfaction drops below Z, we pivot"
- "If costs exceed budget by more than 30%, we reassess"
Write these down. Revisit them. It's harder to ignore a pre-committed decision criterion than to rationalize in the moment.
Separate the Decision from the Decider
If Maria's identity is wrapped up in the CRM project, Maria can't objectively evaluate it. Have someone with no emotional investment assess the situation.
Reframe Stopping as Winning
Killing a bad project isn't failure — it's the smart allocation of scarce resources. The money saved by stopping goes toward something that actually works. That's a win.
Where It Hides
The sunk cost trap isn't limited to big projects:
- Finishing a bad book because you're 200 pages in
- Staying in a bad meeting because you've already been there 30 minutes
- Maintaining legacy code because "we've already built on top of it"
- Keeping a bad hire because "we invested so much in training them"
- Continuing a failing marketing campaign because the budget was approved
The pattern is always the same: past investment is used to justify future investment, regardless of expected future returns.
The Counter-Intuition
The willingness to abandon sunk costs is a competitive advantage. While your competitor pours more money into their failing initiative because they "can't afford to waste" what they've spent, you redirect your resources toward what works. They're optimizing for narrative consistency. You're optimizing for outcomes.
See also: Opportunity Cost | Cognitive Biases | Decision Matrix