What it means
Every manager eventually faces a project that has consumed a large budget and is now underperforming. The question on the table is whether to put in more money or to walk away.
The dilemma arises because the correct answer, based only on future costs and benefits, often feels emotionally wrong. It differs from the related sunk cost trap, which describes the pattern of repeatedly throwing good money after bad.
The dilemma describes the decision point, where a leader must weigh a rational analysis against reputation, team morale, and the wish to avoid reporting a loss. Handled well, it ends with a clear and documented choice.
A useful way to cope is to reframe the question. Instead of asking "how can we justify what we have spent?", ask "if we had not started yet, would we begin this project today on these numbers?" If the answer is no, then the case for continuing relies only on sunk costs.
Uncertainty makes it harder. Future benefits are estimates, and continuing may keep valuable options open, such as a chance of a breakthrough.
Sensible decision-makers therefore weigh probabilities, using expected value, which is the average result across possible outcomes weighted by how likely each one is. Organisations can reduce the pressure with simple controls.
Examples include stage gates, which are scheduled review points with agreed stop criteria, independent reviews by someone who did not start the project, and a culture that treats well-reasoned stopping as a success. Without these, people who championed a project rarely feel able to end it, however clearly the evidence points that way.
In practice
Real-world examples.
Example
A hospital group has spent $3 million on a patient records system that staff dislike. The chief financial officer asks whether the group would buy the system today if it were starting fresh, and the answer is no. The group therefore moves to a cheaper alternative and writes off the old system's remaining book value in the accounts.
Example
A marketing director has spent $120,000 on an influencer campaign that is bringing in few customers. She sets a stage gate with a stop rule and ends the campaign when the next month's cost per customer is above the agreed limit. The saved budget is redirected to a channel with a proven return, and the team is thanked for making a clear decision.
Example
A property developer has paid $2 million for land and found that planning permission is unlikely. He compares the cost of a final application against the chance of approval, and sells the land when the odds are poor, rather than spending more to protect his earlier outlay. He records the lesson in the firm's review notes so that future land purchases are conditional on planning advice.
Formula
Calculation
Expected value of continuing = (probability of success x value if successful) - future cost
Suppose a company has spent $400,000 on a software product. Finishing needs another $300,000, and there is a 40% chance the product succeeds and is worth $700,000. Expected value of continuing = (0.40 x 700,000) - 300,000 = 280,000 - 300,000 = -$20,000. The alternative is to stop and sell the code for $50,000, which has an expected value of +$50,000. Stopping is better by 50,000 - (-20,000) = $70,000, and the $400,000 already spent appears nowhere in the comparison.Case study
Seen in the real world.
Falcon Ridge Brewing is an illustrative, fictional craft brewer that spent $850,000 on a canning line for a new product range. Six months in, sales were a third of the forecast and the line needed a further $250,000 upgrade to meet quality standards.
The founder was torn. Walking away felt like admitting the original plan was wrong, but the finance manager showed that the upgrade would only be worth it if sales doubled, which looked unlikely on current evidence.
Using a simple expected value table, the team found that continuing had an average value of -$60,000, while selling the line to a contract packer would return $200,000. In this illustrative case, the founder sold the line and used the lesson to introduce stage gates on every future investment.
Watch out
Common mistakes.
- Treating the dilemma as a question of loyalty to the project or the team, when it is a question about future value.
- Letting the person who started the project make the stop decision alone, which makes bias more likely.
- Assuming that stopping is always the rational answer, when a project can be worth continuing if its future benefits exceed its future costs.
Questions
People also ask.
How can I tell whether I am facing a sunk cost dilemma?
If your main argument for continuing is "we have already spent so much", rather than a forecast of future returns, you are probably in one.
What is the best single test?
Ask whether you would approve the project today if it were a new proposal with the same future costs and benefits.
Is there a way to take emotion out of the decision?
Yes, agree stop criteria and review dates in advance, and have an independent person review the numbers.
From the founder's library
Take it further with the book.
Build your financial confidence beyond this definition. Shihan's full-length guide, Accounting Fundamentals, takes the same plain-English approach and turns it into a complete, practical playbook for non-finance managers, business owners and students - with chapter-end quiz answers and presentation slides included.
25% off with code MMHQ25, applied at checkout. Priced in USD - checkout may show the equivalent in your local currency.
View the book and save 25%Related