Stopping work is a leadership capability because a new priority often needs people and attention already committed elsewhere. A leadership team can rank one initiative lower without changing its budget, customer promises or place in a specialist’s schedule. The priority above it then has to compete for capacity that has not been released.
The difficult decision is rarely whether an activity has any value. Most work has a credible sponsor and a plausible benefit. Leaders must decide whether its future value justifies what it prevents them from doing, then own the consequences of ending it. A shorter list of priorities will not create focus unless someone settles remaining obligations, moves the resources and defends the decision when requests return.
Why worthwhile work stays on the agenda
An existing initiative has a history that a new priority does not. People have built it, customers may rely on it, and its sponsor can explain what another stage might achieve. Stopping it makes a loss visible to those people. Continuing it spreads the cost across other work through delayed decisions and shared staff. That difference makes postponing the stop decision attractive, even when leaders agree that the portfolio is too full.
Past investment adds another difficulty. Money already spent cannot be recovered by continuing, but the work may have produced useful knowledge. Research by Hun-Tong Tan and J. Frank Yates found that explicit estimates of future returns reduced sunk-cost effects in constructed decision problems. The practical lesson is not to disregard the past indiscriminately. It is to separate money that is gone from evidence that changes what the next investment could produce.
Some commitments also survive a decision to stop new work. A company may still owe customers delivery or support. A team may need time to transfer knowledge or close an agreement responsibly. Unless leaders distinguish those duties from further development, a necessary transition can become an open-ended reason to keep the initiative alive.
OutcomesLab calls the decline in effectiveness caused by spreading finite attention, resources and capability across too many priorities Dilution Drag. A lower ranking does not reduce that drag if the activity still claims staff and decisions. The leadership team has to change what work is authorized, while providing for what remains owed.
Compare the next use of capacity
Peter Drucker connects concentration with the review of existing work in The Effective Executive. His argument is not confined to managing an executive’s calendar: activities that no longer justify their claim on resources should be curtailed so that new work has room to proceed. That puts an established initiative under the same scrutiny as a proposed one.
The proposed OutcomesLab Stop-Work Protocol turns that scrutiny into a practical judgment. Leaders examine whether the activity still fits their strategic direction, what continuing it displaces, whether a limited next stage would preserve a valuable opportunity to learn, and what stopping would make difficult to reverse. They then name the decision owner and the destination of any capacity released. The considerations do not produce a score; they help distinguish work to stop from work to narrow, pause or continue.
Strategic fit alone is not enough. An initiative might support the strategy but be less valuable than another use of the same scarce team. Equally, a project that has missed its original forecast might now have stronger prospects because its team has learned what customers need. The decision concerns expected benefits and costs from here, including the cost of closing the work and the value of what cannot be done while it continues.
That comparison matters especially for a long-horizon bet. Ending it before a planned test could destroy an option worth keeping. But calling an initiative an “option” does not make its demands on staff disappear. Ron Adner and Daniel Levinthal argue that staged investment provides little protection when organizations cannot actually abandon a project; they also caution that rigid stopping boundaries can discard learning. A bounded continuation therefore needs a specific question to answer, a resource limit and a point at which leaders will make another decision.
Consider a hypothetical software company whose engineers develop custom reports for a group of customers. The commercial sponsor can point to useful requests and what the team has learned. Leadership has also made reliability of the main product a priority, and both efforts need the same experienced engineers. The choice is not between valuable and worthless work. It is whether further custom development merits the engineering time that reliability would lose.
On those assumptions, leaders could stop accepting new custom-report requests while completing or renegotiating existing promises. The commercial team could test whether customers want a standard reporting offer, with a defined review date and no standing claim on engineers for new development. That preserves a route to stronger evidence without treating every incoming request as permission to resume expansion. If credible demand and a workable delivery plan later make reporting the better use of those engineers, leadership can change the decision openly.
Make the stop-doing list usable
A stop-doing list should be a record of choices, not an inventory of unpopular tasks. Each entry needs to say what activity will stop or narrow and what outcome it was intended to serve. That distinction matters: ending custom development, for example, does not erase a promise already made to a customer or rule out meeting the same need through a standard offer.
The entry should capture the expected future value of continuing, the competing use of capacity, and any evidence that would warrant reconsideration. It should also identify obligations, dependencies and transition costs; the owner responsible for resolving each; and the executive authorized to approve an exception or restart. These details let teams tell necessary closeout work from unauthorized new work.
Finally, the list must show what will actually be released. Which people, budget and leadership time will become available, when, and for which remaining commitment? If an engineer must continue supporting existing reports part-time, that time is not available for reliability work. If stopping a project saves future spending but requires a closeout payment, the spending decision should reflect both. Closing a budget line while its work migrates to another team is not a completed stop.
A review can then test the decision against assignments and spending rather than count announced cuts. Did outstanding duties move to named owners? Did the engineers reach the work leadership chose to protect? Did a request reopen development without approval? OutcomesLab’s Focus Multiplier describes the disproportionate impact that can arise when scarce resources and attention concentrate on a few coherent choices. Stopping creates the possibility of that concentration; it does not establish that the chosen priority is sound or that the resources have reached it.
Own the decision when work returns
The sponsor of stopped work may have made a responsible case and delivered real value. Leaders should explain the forward-looking trade-off, recognize what the team achieved and state what evidence could change the decision. Calling the initiative a failure when it is being displaced by a more important commitment would obscure the judgment. Leaving the sponsor to explain the stop alone would shift the political cost away from those who made it.
Teams also need a clear boundary they can apply. A customer request may be an existing obligation, genuinely new evidence or simply another request for work leadership has ended. The decision owner should determine which it is and, if an exception is approved, identify what will give way. Otherwise, individually reasonable exceptions can restore an initiative’s claim on capacity without a portfolio decision.
The same discipline applies in the other direction. Leaders should not defend a stop merely to appear consistent. If new evidence materially improves the activity’s future case, they can reconsider it alongside the work it would displace. Until then, the test of saying no is whether remaining obligations are met, the declined work has ended and the chosen commitments receive the capacity leaders said they would have.