Time and money already spent on a project do not show that the next investment is worthwhile. The only useful comparison is forward-looking: what will the next stage cost, what is it likely to produce, and what else could the same people and budget do instead? When past spending becomes the main argument for continuing, the company has fallen into the sunk-cost trap, and the longer it continues, the more expensive the eventual exit becomes.

What is the sunk-cost fallacy?

The sunk-cost fallacy is the tendency to keep investing in something because of what has already gone into it. “We’ve come too far to stop now” sounds like commitment, but the money and months already spent are gone whatever you decide next. They cannot be recovered by continuing, only by stopping sooner.

The trap is psychological rather than analytical. People feel losses more sharply than equivalent gains, a pattern behavioral economists call loss aversion. Ending a project turns a vague disappointment into a definite loss, which feels worse than carrying on. Admitting that a plan was wrong can also feel like a personal failure, especially for the person who proposed it.

Why is it harder to stop in Japan?

Every business culture has this problem. A few features of Japanese business make it more persistent.

  • Consensus decisions are hard to reopen: A plan approved through nemawashi and ringi carries the agreement of many people. Stopping it can feel like telling all of them they were wrong. I look at how to work with that process rather than against it in staying adaptable while doing business in Tokyo.
  • Endurance is admired: Gaman (我慢), bearing difficulty with patience, is a real virtue. Applied to a failing project, it becomes a reason to keep going long after the evidence says stop.
  • Long relationships with vendors: A system integrator or supplier who has served the company for years is hard to replace even when the relationship no longer works. Ending it can feel like a breach of trust.
  • Face: Stopping a project in public can embarrass its sponsor. People avoid raising the question to protect a colleague.

What do zombie projects cost?

Projects kept alive by sunk costs rarely fail outright. They continue in a half-alive state, consuming budget, attention and some of the best people in the company. Staff know the work is unlikely to succeed and still have to report on it, defend it and fit other work around it. Over time that erodes morale and teaches people that nobody will call an end to anything.

The largest cost is what does not get done. Every hour and every yen spent on a project that should have ended is unavailable to one that is working.

In technology, the pattern is especially common. A custom system built by an outside vendor years ago keeps receiving maintenance fees and small change requests because replacing it feels like writing off the original investment. A SaaS tool bought for a project that never took off keeps renewing because nobody wants to admit it is unused. Staff keep maintaining a spreadsheet-based process because a lot of effort went into building it. I cover the cost of the second pattern in what SaaS sprawl costs Japanese SMEs and the first in the hidden cost of good enough systems.

How do you decide whether to stop a project?

Ask the zero-based question

Replace “Should we invest more?” with a question that removes the past from the decision: “If this project were proposed today, knowing what we know now, would we approve it?” If the honest answer is no, or a hesitant maybe, you have your answer. The question works because it does not ask anyone to admit a mistake. It only asks them to judge a proposal on current facts.

Compare the next stage with the alternatives

Estimate what the next stage will cost and what it will realistically produce. Then list what else the same budget and people could do. Most projects look reasonable in isolation. They look different next to the work they are crowding out.

Run a pre-mortem

The pre-mortem, a technique popularized by psychologist Gary Klein, asks the team to imagine it is a year from now and the project has failed, then write down why. Because the failure is hypothetical, people can raise concerns without directly criticising the plan or its sponsor. That makes it useful in hierarchical teams where juniors would not otherwise speak. If the reasons people list are already happening, the decision becomes much easier.

Set stopping conditions in advance

For new projects, agree at the start what would cause you to stop: a milestone missed by a certain margin, a cost ceiling, or a usage level not reached by a certain date. Write it into the approval document. Stopping then follows a rule everyone signed, rather than a judgment that someone has to make against the project’s sponsor.

Make stopping a normal part of management

Talk about sunk costs by name in leadership meetings, so that questioning a project is seen as ordinary diligence rather than disloyalty. When a team decides to end an initiative that is not working, recognize the decision. Thank the people involved for the discipline it took, and make sure their next assignment does not look like a demotion. If ending a project damages careers, people will keep quiet about the next one.

Create a simple process for winding things down. Decide what happens to the data, the contracts and the people, and communicate it clearly to vendors and customers. A tidy exit protects relationships that a messy one would damage, which matters a great deal in Japan.

Finishing the right things

Persistence and loyalty are good qualities, and the sunk-cost trap works by turning them against you. Good leadership is not finishing everything you start. It is finishing the things that still deserve to be finished and releasing people and money from the rest.

If you suspect your company is paying for systems, subscriptions or vendor contracts that no longer earn their place, my Diagnostics engagement reviews your tools, costs and workflows and shows which to keep, replace or retire. When the answer is to replace something, I can help plan and carry out the change without disrupting the work that depends on it.

Further reading: staying adaptable while doing business in Tokyo · turning a business decision into working practice in Japan · Daikin, Elliott and the Japan Inc. value gap