When to Kill a Project (And How to Do It Without Killing Trust)

Killing a project is one of the hardest calls in project management — not because it's technically complex, but because no one wants to say it out loud. This episode breaks down how to recognize when cancellation is the right move and how to do it without destroying team trust.

Show Notes

Every project organization has experienced it: a room full of people who privately know a project should stop, yet the status report stays amber and the meeting ends without the real conversation happening. This episode of ProjectManager.co tackles the uncomfortable discipline of deliberate project cancellation — why it rarely happens soon enough, how to build the process that surfaces the decision, and how to protect the trust of the people who did the work.

Here's what the episode covers:

• Why sunk cost thinking keeps failing projects alive — sponsors who have publicly championed an initiative face real personal pressure to avoid what feels like an admission of failure, even when continued investment makes no rational sense.
• The gap in the stage-gate process — most gate reviews ask whether the project is on track to deliver what was planned, not whether it should still be delivered at all. Those are fundamentally different questions, and only one of them can surface a cancellation recommendation. Revisiting the project lifecycle with that lens changes what gets discussed at every gate.
• A single structured question worth adding to every major review — "If we were making this investment decision today, from scratch, would we make it?" Revisiting the business case template at key milestones rather than treating approval as permanent is a concrete process change any team can adopt immediately.
• How to frame the cancellation recommendation — bring it as a decision document, not a complaint. Restate the original justification, summarize what has changed, quantify the cost of continuing against the benefit that remains, and end with a clear recommendation. Structured project governance gives sponsors and steering groups something to decide rather than something to react to emotionally.
• Protecting team trust after cancellation — teams whose projects are stopped can internalize the outcome as personal failure. Leaders need to explicitly separate the business decision from a verdict on delivery quality, and publicly acknowledge what was genuinely accomplished.
• The governance flywheel — organizations that cancel projects well, early, and with clear reasoning become better at approving the right projects in the first place. It builds a culture where business cases are taken seriously because everyone knows they will be revisited.

For more on how to approach the decisions that sit upstream of delivery, explore the episode on The Dependency Conversation Nobody Has Until It's Too Late — a natural companion to this one. Additional frameworks and templates referenced throughout are available at risk management and across the ProjectManager.co resource library.

ProjectManager.co

What is ProjectManager.co?

Project management for people who deliver, not people who certify. Estimation that survives contact with reality, scope and dependency management, status reporting nobody has to decode, and where self-hosted and AI-assisted tooling genuinely helps.

Each episode takes one practice — a re-estimate mid-project, a dependency nobody owns, a status meeting that produces nothing — and works through a better way to handle it. Written for delivery leads and PMs running real projects, with no methodology evangelism. Five or six minutes, one practice at a time.

Topics include estimation that survives reality, dependency and critical path management, scope negotiation, status reporting nobody decodes, resource planning, retrospectives that change something, and self-hosted versus SaaS tooling.

Produced by ProjectManager.co, self-hosted AI project management. Full details, services and further reading at https://projectmanager.co