When to Kill a Transformation Program

Every large program has a decision nobody wants to make. Not the decision to start, and not the decision to keep going. The decision to stop.

Starting is easy. There is a business case, a sponsor, energy in the room. Continuing is easy too, because continuing is the path of least resistance. Stopping is the one choice that asks a leader to stand up and say the last stretch of work will not deliver what it promised. Most leaders never make it. Failing programs rarely get killed. They get quietly extended until they collapse under their own weight or get absorbed into the next reorg.

The trap of money already spent

The money and time you have already poured into a program are gone. You cannot get them back by continuing, and you do not lose them again by stopping. They are irrelevant to the only question that matters: will the next dollar and the next month produce more value here than anywhere else you could put them?

Economists call this the sunk cost fallacy, and it sounds obvious on paper. In a boardroom it is anything but. Eighteen months and a large budget into a platform migration, no executive wants to be the one who declares it a loss. The spend becomes a reason to keep spending. The longer it runs, the harder it is to stop, because the number that would have to be written off keeps growing.

So the logic inverts. Instead of asking whether the program deserves the next investment, the room asks how to protect the investment already made. Those are not the same question, and confusing them is how organizations fund failure for years.

How dying programs stay alive

A program that should be stopped rarely announces itself. It hides behind activity.

The goalposts move

The original outcome gets quietly redefined. A migration that was going to consolidate systems becomes a project to "stabilize" the systems you already have. Success is restated in terms the program can still hit, not the terms that justified it. Nobody decides this in a meeting. It happens one status update at a time.

The people become the argument

Teams have worked hard. Vendors are embedded. Careers are attached to delivery. Stopping feels like telling all of them their work was wasted, so the human cost of stopping gets weighed against the financial cost of continuing, and the humans win. That is decent instinct and terrible governance.

Adoption gets assumed, not tested

The part that quietly kills most transformations is that the people expected to use the new way of working were never going to. That risk was real at kickoff, and it is usually the first thing to drop off the risk register once delivery pressure sets in. A program can be on time, on budget, and still dead on arrival because nobody will adopt what it built.

What it takes to stop one

Killing a program well is not an act of nerve. It is a discipline you build before you need it.

Name the outcome, not the milestones. Write down, in plain language, the business result that justified the program. Return to that sentence at every gate. If the honest answer to "can we still get there" is no, milestones are noise.

Separate the people from the decision. The team that built the thing cannot be the sole voice on whether the thing should live. That is not a comment on their integrity. It is a recognition that no one can assess their own work without bias. Bring a voice to the table that has nothing invested in the answer.

Make stopping a normal option, not a scandal. If the only outcomes a steering committee ever considers are green, amber, and delayed, it is not governing. Every serious review should include an explicit option to stop, with a plan for what happens to the people and the budget if it does. When stopping is on the menu, it stops being a career-ending admission.

Say the sunk cost out loud, then set it aside. Acknowledge the spend. Do not pretend it did not happen. Then state clearly that it is not a factor in the decision ahead, and hold the room to that. Naming the trap is how you disarm it.

Stopping is a result

There is a version of leadership that treats a killed program as a mark of failure. I see it the opposite way. Reallocating money and talent away from something that will not pay back, and toward something that will, is one of the highest-value moves an executive can make. It frees people who were being burned on a lost cause. It restores credibility to the next thing you ask them to believe in.

The organizations that get transformation right are not the ones that never make a bad bet. They are the ones that can call a bad bet early, cut it cleanly, and move. That takes a table where stopping is a legitimate answer and someone with the standing to say it.

If your steering committee has never seriously debated stopping, that is not proof the program is healthy. It is proof no one is asking the question.