Using Stage Gates to Stop a Failing Project Without Blame

## The short answer

The most valuable thing a stage gate can do is **stop a failing initiative early** — and it only does this well when stopping is treated as a good outcome, not a personal failure. If your boards never say no, your gates are decorative. The fix is partly process (clear stop criteria, sunk-cost discipline) and partly culture (rewarding honest "stop" decisions). Done right, killing a weak idea at gate two is a success, because it frees money and people for something better.

## Why good projects are so hard to stop

Several forces conspire to keep failing work alive:

- **Sunk-cost thinking** — "we've spent so much already" feels like a reason to continue, when the money already spent is irrelevant to whether the *next* pound is worth spending.

- **Optimism and momentum** — teams genuinely believe the next milestone will turn things around.

- **Reputational risk** — sponsors fear that stopping signals their judgement was wrong.

- **No natural off-ramp** — without explicit stop points, continuing is the default and inertia wins.

Stage gates exist precisely to interrupt this inertia with a scheduled, legitimate moment to ask: knowing what we know now, would we start this today?

## The single most useful gate question

Reframe every gate around a forward-looking test:

> *Given everything we now know, and ignoring what we have already spent, would we choose to fund the next phase of this initiative from scratch?*

This question neutralises sunk cost. If the answer is no, the gate should stop or pivot the work — regardless of how much has been invested.

## Define stop criteria, not just pass criteria

Most gate frameworks list what must be true to *proceed*. Add the inverse — conditions that should trigger a stop or serious pivot:

- The original problem has changed or been solved another way.

- Expected benefits have materially shrunk or become unprovable.

- Costs or risks have grown beyond the appetite agreed at funding.

- A key assumption underpinning the business case has been disproven.

- A better-value initiative now competes for the same resources.

Naming these in advance makes stopping a *pre-agreed* response to evidence, not a surprise judgement on people.

## Make stopping culturally safe

Process alone will not save you if stopping feels like career risk. Build the culture deliberately:

- **Celebrate good stops.** Publicly recognise teams that surfaced bad news early and freed resources. Frame it as portfolio hygiene, not failure.

- **Separate the idea from the people.** A stopped initiative reflects on the evidence, not the competence of those who ran it.

- **Reward honesty over heroics.** Punishing the messenger guarantees future bad news arrives too late to act on.

- **Tell the stop stories.** Organisations that only narrate successes teach everyone to hide failure until it is unstoppable.

## Use the portfolio lens

A stop decision is easier when framed as reallocation. The question is rarely "is this initiative bad?" but "is this the best use of these people and this money right now?" Viewing gates at portfolio level — where stopping one thing visibly funds another — turns a negative into an obvious trade. Enterprise delivery platforms, including those neart.ai builds, make this trade visible by connecting initiatives to benefits and resources, so a stop reads as a reallocation rather than a loss.

## Handle the mechanics with care

Stopping well is itself a small project:

- **Capture learning** before the team disperses — what the evidence showed and why.

- **Salvage value** — reusable components, validated insights, data.

- **Redeploy people** quickly and visibly, so stopping does not feel like punishment.

- **Communicate clearly** to stakeholders, framing the decision as disciplined investment management.

## Avoid the pivot-as-disguise trap

Beware the "pivot" that is really a refusal to stop. A genuine pivot has a fresh, credible hypothesis and a reset of expectations; a fake pivot relabels the same failing work to dodge the difficult decision. Hold pivots to the same forward-looking funding test as any new initiative.

## Takeaway

Stage gates earn their keep by stopping weak initiatives early. Ask at every gate whether you would fund the next phase from scratch today, ignoring sunk cost; define explicit stop criteria; and build a culture where an honest, well-evidenced "stop" is celebrated as freeing resource for better work. A governance process that never says no is not governance at all.