Short answerThree cognitive biases quietly wreck founders' money decisions: the sunk cost fallacy (throwing good money after bad), loss aversion (a loss hurts about twice as much as an equal gain feels good, per Kahneman and Tversky), and overconfidence (systematically underpricing risk). Naming them in the moment, and running a short pre-decision checklist, is the cheapest debiasing tool you have.
You already know the founder who kept pouring money into a dying product line because they'd "come too far to stop." Behavioural biases in financial decisions aren't a personality flaw; they're the default wiring of every brain under uncertainty, documented in decades of Nobel-winning research. Here are the three that cost founders most, each with an example and a tactic to catch it.
Which cognitive biases affect founders' financial decisions?
Dozens do, but three explain most self-inflicted financial damage: sunk cost, loss aversion and overconfidence. Each makes the emotionally easy choice feel like the rational one. Daniel Kahneman and Amos Tversky showed these aren't rare errors by irrational people; they're systematic and universal, mapped for general readers in Kahneman's Thinking, Fast and Slow (2011).
1. The sunk cost fallacy: throwing good money after bad
The sunk cost fallacy is the pull to keep investing because of what you've already put in, rather than what you'll get back. Harold Arkes and Catherine Blumer demonstrated it cleanly in 1985: theatregoers who had paid full price for a season ticket attended more plays than those given the same ticket at a discount, even though the price paid was gone either way (Arkes & Blumer, The Psychology of Sunk Cost, 1985).
For founders it's the feature nobody uses that you keep staffing because it took a year to build, or the channel you keep funding because you've already burned $500,000 there. The money is gone; the only question that matters points forward: from today, is this the best use of the next dollar?
Debiasing tactic: reframe every continue-or-kill call as a fresh buy. Ask, "Knowing what I know now, if I hadn't started this, would I begin it today at this cost?" If not, you're funding a sunk cost, not a future.
2. Loss aversion: why a loss hurts twice as much
Loss aversion is the reason a $10,000 loss stings more than a $10,000 gain delights. Kahneman and Tversky's prospect theory (1979) established that "losses loom larger than gains," and their later work put a number on it: the pain of a loss runs to roughly twice the pleasure of an equivalent gain, with the loss-aversion coefficient estimated near 2 (Tversky & Kahneman, Advances in Prospect Theory, 1992).
In practice, loss aversion makes founders hold losing bets too long and cut winning ones too early. You keep a failing product on life support because closing it "locks in" the loss. It distorts pricing too: founders underprice for fear of losing a customer, when a few churned accounts are worth the margin.
Debiasing tactic: convert the decision to absolute numbers. Strip the words "loss" and "gain" and look only at the two future balances: "Keep it: minus $8,000 a month. Kill it: one-time $20,000, then zero." Numbers don't trigger the flinch the word "loss" does.
3. Overconfidence: mispricing your own odds
Overconfidence is the systematic tendency to think our estimates are more accurate, and our odds better, than they are. Colin Camerer and Dan Lovallo showed its economic bite in 1999: when experimental entrants competed on skill, too many entered and the group lost money, because each rated their own chances above average (Camerer & Lovallo, Overconfidence and Excess Entry, 1999).
For founders, overconfidence hides in the model: revenue ramps that only bend upward, a raise that will "definitely close in eight weeks," a single scenario where everything goes right. It's why runway forecasts skew optimistic and why founders sign leases against revenue they don't yet have.
Debiasing tactic: take the outside view. Instead of "how long will our raise take?", ask "how long has it taken comparable startups?" Then build three cases, not one, and pressure-test the downside, the discipline behind downturn scenario planning and honest startup runway math.
Honourable mentions: confirmation and anchoring
Confirmation bias makes you seek data that supports what you've already decided. Anchoring, documented by Tversky and Kahneman in 1974, lets the first number set the range: an investor's opening valuation becomes the reference point the whole round orbits. If one number is steering you, ask where it came from.
A checklist to catch these behavioural biases in the moment
Before any big financial decision, run these five questions. Two minutes interrupts the autopilot.
- Fresh-start test (sunk cost): If I were deciding today with no history, would I still choose this?
- Balance test (loss aversion): Have I compared the two future outcomes in absolute numbers, with "loss" and "gain" removed?
- Outside view (overconfidence): What actually happened to comparable companies, not what I hope happens to us?
- Disconfirming evidence (confirmation): What would change my mind, and have I looked for it?
- Anchor check (anchoring): Is one number driving this, and where did it come from?
Write your answers down. Writing forces the deliberate thinking Kahneman calls System 2 to override the fast, emotional default.
What this means for you
You can't delete these biases; Kahneman said studying them for decades barely improved his own snap judgements. What works is friction: a checklist, a second opinion, a 24-hour wait on big decisions. Founders who build one debiasing habit into spending, usually the fresh-start test, stop repeating the same expensive mistake. The same wiring shows up in treasury, which is why where to park your startup's cash is as behavioural as it is financial.
Frequently asked questions
What's the difference between the sunk cost fallacy and loss aversion?
Sunk cost is about the past: you keep investing because of what you've already spent. Loss aversion is about the future: you fear a loss more than you value an equal gain. They combine, because closing a project feels like locking in a loss, which feeds the sunk-cost pull to keep funding it.
Are entrepreneurs really more overconfident than everyone else?
On balance, yes. Starting a company at all requires believing your odds beat the base rate. Camerer and Lovallo's 1999 work showed that people who back their own skill enter markets in excess, which helps explain high failure rates. Unchecked overconfidence in your financial model is the dangerous kind.
How do you actually make less biased financial decisions?
Add friction to big calls. Run a short pre-decision checklist, force the numbers into absolute terms, seek one piece of disconfirming evidence, and ask what happened to comparable companies. A 24-hour cooling-off rule and a second reviewer catch most bias in the moment, the only place it can be caught.
What is loss aversion in one sentence?
Loss aversion is the well-documented tendency for a loss to feel about twice as painful as an equivalent gain feels good, from Kahneman and Tversky's prospect theory. It pushes people to hold losing positions too long and take too little risk on winners, which is why founders keep failing projects alive.
Sources
- Kahneman, D. & Tversky, A. — Prospect Theory: An Analysis of Decision under Risk, Econometrica 47(2), 263–291, 1979. https://www.jstor.org/stable/1914185
- Tversky, A. & Kahneman, D. — Advances in Prospect Theory: Cumulative Representation of Uncertainty, Journal of Risk and Uncertainty 5, 297–323, 1992. https://link.springer.com/article/10.1007/BF00122574
- Arkes, H. R. & Blumer, C. — The Psychology of Sunk Cost, Organizational Behavior and Human Decision Processes 35(1), 124–140, 1985. https://www.sciencedirect.com/science/article/abs/pii/0749597885900494
- Camerer, C. & Lovallo, D. — Overconfidence and Excess Entry: An Experimental Approach, American Economic Review 89(1), 306–318, 1999. https://www.aeaweb.org/articles?id=10.1257/aer.89.1.306
