An early win deserves further investment when it reveals a way to make the next result more reliable. A favorable number is a starting point: leaders still need to understand what produced it, whether the method can travel and which decision would allow others to repeat it.
Consider a hypothetical service pilot that records fewer repeat customer requests. If resolving a common exception sooner explains the improvement, extending frontline authority could help the next team. If the number fell because of overtime or changed recording, expansion would spread a fragile practice. The distinction determines what leaders should support next.
Make the result worth trusting
For this hypothetical pilot, “improve customer experience” does not tell staff which problem to solve. The team needs to name the recurring issue and the improvement customers should experience. It also needs agreement from anyone whose help is essential. If frontline staff cannot close a case without technical support, that team must know what it has committed to do. Otherwise, the pilot becomes an unplanned interruption to its other work.
John Doerr’s Measure What Matters is associated with objectives and key results: stating what a team wants to achieve, then identifying evidence of progress. Google’s guidance on the approach calls for evidence others can inspect and for dependent teams’ contributions to be explicit. A target will not settle the handoff between service and technical support. It can make that handoff visible as a decision leaders need to make.
Before the pilot starts, the team could agree to measure the share of customers who return with the same issue within a defined period. Leaders would need a starting level for comparison and a consistent way to recognize a return request. They would also need to ask what else might move the number. Tickets closed might show that staff are busy; they would not answer whether customers have stopped needing help twice. Agreeing on the measure in advance makes it harder to substitute activity for the intended result.
A credible result can matter to the people doing the work, too. Teresa M. Amabile and Steven J. Kramer’s research links perceived progress on meaningful work with employees’ day-to-day motivation. Seeing customers get an issue resolved may encourage the service team to keep improving its method. It does not establish that the method will work elsewhere. Colleagues first need to understand what produced the result.
Use proof to change the work
Suppose repeat requests remain lower and customer feedback supports the result. In a review with frontline, technical-support and operations leaders, the team finds that resolving one common exception sooner appears to account for much of the improvement. The next question is specific: who can approve that resolution during ordinary service, and what help will staff need to do it reliably?
The proposed OutcomesLab sequence is the Strategic Momentum Flywheel. Evidence from the pilot gives people a reason to commit to the next step. Leaders settle an approval rule or protect technical-support time. That change gives staff a better chance of resolving the next customer’s problem without exceptional help. The next result then gives leaders something to test again. The flywheel is a way to think through those decisions, not a promise that one win will produce another.
Peter M. Senge’s account of reinforcing feedback in systems thinking helps explain the possibility. Senge, author of The Fifth Discipline, describes how a result can prompt further investment and then another result. He also notes that delays and conflicting goals can change the pattern. In the service pilot, a favorable customer measure cannot grant staff authority, train the next shift or make room in technical support’s schedule. Leaders still have to remove the constraint the pilot has revealed.
That decision need not require a larger budget. Leaders might protect technical-support time by stopping a lower-priority report, or revise an approval rule and teach the method to another team. When each round leaves a more reliable method, capability compounds. Momentum concerns the further step: a credible result changes the commitment that makes the next round possible. Repeated overtime or executive rescue provides neither a dependable method nor a sustainable commitment.
Even a sound method can fade when it conflicts with ordinary management. Suppose a review praises fewer repeat requests, but the standing performance measure still rewards rapid ticket closure regardless of whether customers return. Staff now have conflicting instructions. Unless the person who owns that measure joins the decision, the pilot remains an exception rather than a change in how service works.
Check whether the loop deserves to continue
The same process that builds momentum can protect a mistake. Once a result wins attention and resources, people have more reason to defend it. Lisa D. Ordóñez and her coauthors warn that narrow goals can draw attention away from work the goal does not measure. That is a risk to manage, not a reason to abandon a useful target.
Fewer recorded repeat requests would be misleading if customers gave up, went elsewhere for help or waited longer for their first resolution. Before expanding the pilot, leaders should check whether customers say the issue was resolved, whether resolution time or the technical-support queue worsened, and whether another team can use the method without extraordinary assistance. If those checks weaken the original story, change the pilot or reconsider the strategy rather than reinforce the headline number.
Give the pilot a regular review with authority to make one practical decision. Start with what changed for customers and how confident the team is about the cause. Then decide which resource, routine or dependency to change for the next round—or what to investigate before extending the pilot. Progress becomes self-reinforcing when that decision makes good work easier to repeat, while leaving leaders free to stop when the evidence no longer supports it.