Sunk Cost Fallacy

Quick Definition

The sunk cost fallacy happens when a founder or team continues investing resources into a product, feature, or campaign because of what has already been spent, rather than evaluating it on its future return. The money, time, or effort already gone is irrecoverable no matter what happens next, but the brain treats it as a reason to keep going.

In startups, this shows up constantly: the feature nobody uses but took three months to build, the ad channel that never converted but already ate half the budget, the co-founder relationship that isn't working but you've been at it for two years.

Why It Matters for Early-Stage Founders

Startups run on scarce resources. Every hour and dollar spent defending a dead idea is an hour and dollar not spent on something that could actually work. The sunk cost fallacy is one of the quietest killers of early-stage companies because it doesn't feel like a mistake in the moment. It feels like persistence, loyalty, or discipline.

The danger is compounding. A founder who can't kill a bad feature also struggles to kill a bad hire, a bad pricing model, or a bad market bet. Left unchecked, sunk cost thinking turns into a company-wide habit of throwing good money after bad.

How It Works

The fallacy operates through a simple, flawed logic chain:

1. We invested X into this.
2. Quitting now means X was 'wasted.'
3. Continuing might 'save' that investment.
4. Therefore we keep going.

The error is in step 3. Past investment cannot be recovered by future spending. The only question that matters is: given where we are right now, is this the best use of our next dollar and next hour?

A Concrete Example

A startup spends 4 months and $60,000 building an in-app referral feature. Usage after launch is under 2% of active users, way below the 15% benchmark the team expected. The team debates: 'We already spent $60k, let's add one more integration to make it work.'

The correct question isn't "how do we protect the $60k already spent." It's: "If we had $0 invested today, would spending the next $15,000 and 3 more weeks on this feature be our best option?" If the answer is no, the $60,000 is gone either way. The only live decision is what happens with the next dollar.

The Reframe Formula

Use this simple test before continuing any struggling initiative:

Future Value Test = Expected future return of continuing vs. expected future return of the next best alternative use of the same resources

If the next best alternative (a new feature, a different channel, or just cutting burn) beats the expected return of continuing, stop. Ignore everything already spent.

Common Mistakes

• Confusing sunk cost with valid signal. Not every underperforming feature is dead. Check whether the problem is execution (fixable) or demand (not fixable) before killing something.
• Letting ego drive the decision. Founders who publicly championed an idea often keep funding it to avoid admitting they were wrong. Separate identity from strategy.
• No pre-set kill criteria. Without clear metrics defined before you launch, it's easy to keep moving the goalposts to justify continued investment.
• Treating team morale as a reason to continue. Killing a project respectfully is usually better for morale long-term than dragging out a slow failure.
• Ignoring opportunity cost. The real cost of a bad feature isn't just its budget, it's what the team could have built instead.

Benchmarks and Practical Guardrails

• Set kill criteria before you build: define the specific usage, revenue, or retention number that means "stop" before you start, not after you're emotionally invested.
• Run a quarterly portfolio review: list every active initiative and force a rank by expected future ROI, not past spend.
• Use a "fresh eyes" test: ask an advisor or someone outside the project, "If we hadn't spent anything yet, would you fund this today?"
• Track a kill rate: healthy startups kill 20 to 30% of features or experiments within 90 days of launch. If your kill rate is near zero, you're likely sunk-cost trapped.

Founders using tools like welaunch.sh to plan and launch new features can build kill criteria directly into the launch checklist, so the decision to continue or cut is made with data before ego and past spend get involved.

The fastest way to protect a startup's limited runway is to treat every past dollar as gone the moment it's spent, and every future dollar as the only one that matters.