Sunk Cost Fallacy: Why Founders Can't Let Go (And How to Fix It)

What Is the Sunk Cost Fallacy?

The sunk cost fallacy is the tendency to continue investing in something because of the resources you've already committed — even when the evidence says you should stop.

The logic goes: "We've already spent $200K and 8 months on this. We can't stop now." But the $200K and 8 months are gone regardless of what you do next. They're sunk. The only rational question is: "Looking forward, is continuing the best use of our resources?"

This seems obvious when stated abstractly. In practice, it's one of the most common and most expensive cognitive errors founders make.

Why Founders Are Especially Susceptible

Identity Investment

For a founder, the company isn't just a financial investment — it's an identity. You've told investors, friends, family, and yourself that this is the thing. Walking away from a product line, a strategy, or even a single feature can feel like admitting that you were wrong, not just that the decision was wrong.

This identity-level attachment makes sunk costs feel visceral rather than abstract.

Public Commitment

Founders make public commitments — to investors, to customers, to the press. Changing course after a public commitment feels like breaking a promise, even when it's the strategically correct move.

The higher your profile, the stronger this effect. The founder who announced the feature at a conference feels more locked in than the one who discussed it in an internal meeting.

Narrative Consistency

Humans are storytelling creatures. We want our decisions to form a coherent narrative. Abandoning a project that you championed disrupts the narrative of "visionary founder who sees what others can't." Continuing it, even when the data is bad, preserves the story.

Loss Aversion

Kahneman and Tversky showed that losing $100 feels roughly twice as bad as gaining $100 feels good. When you frame "stop investing in X" as a loss (of the time and money already spent), the psychological cost is amplified.

Reframing "stop investing in X" as "redirect resources to something better" — a gain — reduces the psychological barrier. But reframing requires conscious effort.

Where Sunk Cost Shows Up for CEOs

The Feature Nobody Uses

You spent 4 months building an enterprise analytics dashboard. Usage data shows 3% of customers have opened it. The team that built it is proud of it. The roadmap assumed it would drive upsell.

Sunk cost response: "We just need to market it better. Let's build onboarding tutorials and push notifications." (More investment in a losing bet.)

Forward-looking response: "The data says customers don't value this. What would we build instead with those engineering resources?" (Evaluate future, not past.)

The Hire Who Isn't Working

You spent 3 months recruiting, negotiated a premium comp package, relocated someone across the country. Three months in, it's clearly not working.

Sunk cost response: "Let's give it another quarter. We invested so much in getting them here." (The investment is already spent. Another quarter of poor performance costs more.)

Forward-looking response: "Is this person likely to succeed in this role? If I were hiring today, would I hire them again? If not, what's the kind thing to do?"

The Pivot You Can't Make

Your original product thesis isn't working, but adjacent opportunities are emerging from customer conversations. The evidence points toward a pivot.

Sunk cost response: "We've raised $2M on this thesis. We have 18 months of code built for this direction. We owe it to our investors to see it through."

Forward-looking response: "Our investors invested in the team and the outcome, not the specific product. The best way to honor their investment is to pursue the highest-probability path to a return."

The Partnership That's Draining Resources

Six months ago, you signed a strategic partnership. It required significant integration work, a dedicated team, and management attention. The partner hasn't delivered on their commitments. ROI is negative.

Sunk cost response: "We've already built the integration. Let's give it one more quarter."

Forward-looking response: "What is this partnership costing us in opportunity cost? Would we sign this deal today? If not, what's the exit plan?"

A Framework for Killing Sunk Costs

The "Fresh Eyes" Test

Ask yourself: "If I were a new CEO starting today, with no history with this project, would I invest in it?"

This question strips away the emotional weight of past investment and forces a pure forward-looking evaluation.

The Opportunity Cost Audit

For any project you're tempted to continue out of sunk cost attachment:

1. What is this project costing per month? (People, dollars, attention)
2. What is the best alternative use of those resources?
3. What is the expected return on continued investment?
4. What is the expected return on the alternative?

If the alternative wins, the sunk cost argument is costing you the difference.

The Pre-Commitment Kill Criteria

Before starting any major initiative, define the conditions under which you'll stop:

• "If we don't see X metric by month 3, we'll pause and reassess"
• "If customer adoption is below Y after the beta, we'll kill it"
• "If this hire doesn't meet these specific milestones by month 4, we'll part ways"

Pre-committing to kill criteria removes the in-the-moment emotional decision. You've already decided what "stop" looks like — now you just need to execute on the plan you made when you were thinking clearly.

The Admiration Reframe

Instead of viewing abandonment as failure, reframe it as a signal of strong leadership:

• Weak leadership: Continuing to invest in losing bets because changing course feels uncomfortable
• Strong leadership: Making clear-eyed decisions about resource allocation, even when it means admitting a previous decision was wrong

The CEOs I most admire aren't the ones who never made bad bets. They're the ones who recognized bad bets quickly and redirected resources decisively.

When Persistence Is Right

Sunk cost awareness shouldn't turn you into someone who quits at the first sign of difficulty. Not everything that's hard is wrong.

Continue if:

• You have a clear thesis for why things will improve, based on new evidence
• External conditions have changed and the original evaluation is no longer valid
• You're in a known "valley" that successful precedents have navigated
• The remaining investment to reach a decision point is small

Stop if:

• Your only reason for continuing is the investment already made
• The original thesis has been clearly disproven
• You wouldn't start this project today with current knowledge
• Continuing is preventing you from pursuing better opportunities

Key Takeaways

1. Sunk costs are gone — only future costs and benefits should drive decisions
2. Founders are especially vulnerable due to identity investment, public commitment, and loss aversion
3. Use the "fresh eyes" test: would a new CEO invest in this today?
4. Set kill criteria before starting any major initiative
5. Cutting losses is a sign of strong leadership, not failure

In Coaching

Sunk cost conversations are some of the hardest in coaching — because you're essentially asking someone to admit that a decision they're emotionally invested in isn't working. The ego wants to prove it can make this work. The business needs someone willing to look at the numbers honestly.

This is where the confidential, non-judgmental nature of coaching matters most. You need a space where you can say "I think this isn't working" without fear of looking weak. That admission is the first step to making the forward-looking decision that the business actually needs.