Most businesses have some way of starting projects: an idea is discussed, a budget is approved, someone is put in charge. Very few have an equally clear way of stopping them. Starting is visible and optimistic. Stopping feels like admitting a mistake. The person who championed the project has their reputation attached to it, the team has worked hard, suppliers are engaged and money has already been spent. Uncertain future benefits are easier to defend than the quieter value of releasing people for something better.
So projects accumulate. New work is added while old work continues at half pace. The same few people are spread across too many things. Strategy changes in conversation but not in the project list. And the business slowly fills with commitments that made sense when they were approved but no longer do.
This article explains why stopping well is a management capability rather than an admission of failure, what research suggests about the quality of stopping decisions, how to set exit triggers in advance, the difference between stopping, pausing and redesigning, and what must be handed over when a project ends so that its value is not lost.
Stopping is how a business renews its priorities
A business that can add projects but cannot remove them will collect past decisions faster than it can act on new priorities. The effects are predictable:
- scarce people stay tied to low-value work;
- stronger opportunities wait;
- the number of active projects stays high;
- changes in strategy remain talk rather than action;
- money already spent starts to drive further spending.
The question is not whether a project is “bad”. It is whether continuing it is still the best use of the business’s people, money and attention compared with the alternatives.
What the research suggests
A study by Unger, Kock, Gemünden and Jonas followed 54 firms, each running at least 20 projects at once, and examined what they called termination quality: how well inappropriate projects were recognised and stopped. Better termination quality was associated with better strategic fit across the firms’ project portfolios. In other words, staying aligned with strategy depends partly on how well a business removes work that no longer fits, not only on how well it chooses new work.
The study also found that senior management involvement helped termination quality up to a point, beyond which more involvement was associated with worse decisions. Leaders who sponsor projects can become attached to them, overestimate their ability to rescue them or protect favourites. The lesson is not that owners should step back from these decisions. It is that accountability and over-involvement are different things. The study involved larger organisations, but the pattern is easy to recognise in small businesses where the owner champions every initiative.
Common mistakes
- Treating a stop as proof the original decision was wrong. A sensible investment can become unattractive when conditions change. A pilot that answers its question has done its job.
- Waiting until failure is undeniable. By then, most of the avoidable cost has been spent. The better moment is often when the project is still recoverable but no longer justified.
- Asking whether it can be finished. Almost any project can be finished with enough time and money. The question is whether finishing it is the best use of those resources.
- Letting the champion be the only judge. The person who sponsored a project is valuable for protecting it, and least well placed to judge when to let it go.
- Leaving projects half-alive. An open-ended “pause” can do more harm than a clear stop: it ties up attention, keeps suppliers waiting and stops people being reassigned with confidence.
Set exit triggers before you need them
Stopping is far easier when the conditions for reconsidering a project are agreed at the start, before anyone is defending it. Possible triggers include:
- the goal the project serves has changed;
- a critical assumption has proved wrong;
- confidence in the benefits has fallen below an agreed level;
- the cost to complete has risen beyond what the business will accept;
- something it depends on has failed;
- a better alternative has appeared;
- the people it needs are needed more elsewhere;
- a pilot has answered the question it was set up to answer.
A trigger should not mean automatic cancellation. It should mean a genuine review of whether the project still deserves its resources. The when the project succeeds and the strategy fails article describes a continuing justification test that fits well here.
Stop, pause, redesign, accelerate or continue
A review should end with one of five clear outcomes:
| Outcome | When it fits |
|---|---|
| Stop | The investment no longer deserves to continue |
| Pause | A specific question must be answered first, with a set decision date |
| Redesign | The goal is still valuable but the current approach is weak |
| Accelerate | Evidence has strengthened the case and delay is costly |
| Continue | The original reasoning still holds |
Having all five options makes it easier to act, because stopping is not the only response to bad news. It also stops “pause” becoming a polite way of avoiding a decision. A pause without a question and a date is not a pause.
Review the least justified projects first
When reviewing a list of projects, the instinct is to start with the one most visibly in trouble. A better starting point is the projects whose purpose is least clear: the oldest, the most often deferred and the ones nobody can link to a current priority. A troubled project may still be very valuable. A smoothly running project that no longer matters is easier to miss and just as costly.
End projects properly
A good stopping decision can still destroy value if the ending is handled badly. Larger change programs offer a useful lesson here: they can end in several ways, and each needs different handling.
- Complete: the intended result is in place and the project structure is no longer needed.
- Redirect: the need remains, but the approach or target changes.
- Transfer: the remaining work belongs in normal operations or a separate smaller task.
- Terminate: the reason for the project has weakened enough to stop, with any useful parts kept.
Whichever applies, check what must survive the project:
- Benefit ownership: who will make sure the expected benefits actually arrive, and how they will be measured. Delivery rarely equals benefit.
- Capability: whether the people who will run what was built have the skills, procedures, spare parts and support they need.
- Obligations: contracts, warranties, commitments to customers and regulatory obligations that continue after the project ends, each with an owner. The how contracts end article covers supplier contracts.
- Knowledge: what was learned about customers, suppliers, technology and the business itself, recorded where the next project will find it.
- People: where the team goes next.
Make sure freed resources go somewhere better
The point of stopping a project is to redirect its people and money to more valuable work. Track where they actually go. If released capacity is immediately absorbed by equally weak work, nothing has improved. Treat each stop as a decision about where those people should work next.
Build a culture where stopping is respected
People will not raise concerns about a project if stopping is treated as failure. Recognise good stopping decisions: the pilot that answered its question quickly, the manager who said a project no longer fitted, the team that redirected effort when evidence changed. Over time, this makes it normal to ask whether a project still deserves its resources, and to answer honestly. The reviews that ask why article covers how to learn from ended projects.
Who should decide
Continuation decisions work best when the person sponsoring a project is not the only one judging it. In a small business, that might mean the owner reviewing projects with one or two managers who are not directly involved, or asking an adviser or board member to challenge the list once a quarter. The sponsor should present the case and the evidence; the decision should rest with someone who can weigh the project against everything else competing for the same people and money.
It also helps to separate the review from the moment of crisis. Decisions made in the middle of a problem tend to favour rescue. A regular, scheduled review of all projects, using the same questions each time, makes stopping a normal outcome rather than an emergency measure.
A worked example
This is an illustration. A manufacturing business with 50 staff has nine improvement projects under way. Two engineers and the operations manager are involved in almost all of them, and progress on each is slow. The owner reviews the list, starting not with the most troubled project but with those least clearly linked to current priorities.
Each project is asked the same questions: does it still support a current priority, do its benefits still justify its remaining cost and people, what has changed since it was approved, and what else could those people do?
The outcomes:
- Three projects are stopped. A warehouse barcode trial has already shown that the main benefit depends on a system upgrade planned for next year; its learning is recorded and it ends. A redesign of a product line that is now in decline is stopped. A supplier portal is stopped because the main supplier has introduced its own.
- One project is paused until the result of a customer’s tender is known in eight weeks, with a decision date in the diary.
- One is redesigned: an energy monitoring project is narrowed from the whole site to the two largest machines.
- One is accelerated: a changeover improvement on the main line, which has shown strong early results.
- Three continue as planned.
Stopping the three projects releases roughly a day and a half a week of engineering time and about $60,000 of remaining budget. The owner deliberately assigns the time to the accelerated changeover project, rather than letting it disappear into other work. For each stopped project, a short closing note records what was learned, what was handed over and who owns any remaining obligations.
How this applies to a small Australian business
Small businesses often have more projects than people, and the owner is usually the sponsor of most of them. Practical steps:
- List every active project in one place.
- Agree exit triggers when each project starts.
- Review the least justified projects first, not just the most troubled.
- Use all five outcomes: stop, pause with a date, redesign, accelerate or continue.
- Involve someone other than the sponsor in continuation decisions.
- Close projects properly: benefits, capability, obligations, knowledge and people.
- Redirect freed resources deliberately.
- Recognise good stopping decisions.
Signals worth watching
- More projects started than finished.
- The same people named on most projects.
- Projects nobody can link to a current priority.
- Pauses with no decision date.
- Spending justified mainly by what has already been spent.
- Freed-up people quietly absorbed into low-value work.
Common mistakes
- Having no process for stopping.
- Waiting for undeniable failure.
- Letting champions judge their own projects alone.
- Using open-ended pauses to avoid decisions.
- Stopping without closing properly.
- Failing to redirect released people and money.
Frequently asked questions
How often should we review whether projects should continue? At major milestones, when an exit trigger occurs and at least every quarter for the whole list.
What if a stopped project’s sponsor is upset? Explain the reasons, recognise their contribution and the learning gained, and involve them in deciding where the freed resources go. Stopping for good reasons, explained well, builds trust.
Isn’t stopping wasteful when money has already been spent? Money already spent cannot be recovered by continuing. Only future costs and benefits matter. Continuing a weak project wastes the future spending too.
What should a closing note include? Why the project ended, what it delivered, what was learned, what was handed over and to whom, any remaining obligations and their owners, and where the team is going next.
How do we avoid stopping projects too readily? Use agreed triggers and evidence, consider redesign and pause as alternatives, and give projects that are exploring something new a fair chance to answer their question before judging them.
What about projects we are contractually committed to? Check the contract’s termination terms and the cost of ending early, and compare that with the cost of continuing. Sometimes a negotiated exit or a reduced scope is better than completion. Take advice before ending significant contracts.
How many projects should a small business run at once? Fewer than most try to. A practical limit is set by the people who are needed on every project, often the owner and one or two key staff. If they cannot give each project real attention every week, there are too many.
Questions to ask
- Which of our projects would we approve again today?
- Which have no clear link to a current priority?
- Who, other than the sponsor, judges whether each project should continue?
- Which pauses have no decision date?
- Where did the people from our last stopped project go?
- Do we treat stopping as failure or as good management?
Bringing it together
Strategy is enforced not only by approving the right work but by withdrawing resources from work that no longer deserves them. Agree exit triggers in advance, review the least justified projects first, use all five outcomes rather than defaulting to “continue”, and involve someone other than the sponsor. When a project ends, hand over benefits, capability, obligations and knowledge, and redirect the freed people deliberately. A business that stops well has more capacity for the work that matters.
Source: KEVOS notes, drawing on research by B. N. Unger, A. Kock, H. G. Gemünden and D. Jonas on project termination quality and portfolio strategic fit, and teaching material on program closure. Examples and figures in this article are illustrations.
