A recurring meeting earns its time through the work it enables. It may reveal a risk, improve someone’s judgment or settle a trade-off across functions. Its purpose determines what people should prepare, who needs to attend and what must happen after the conversation.
Consider a hypothetical executive meeting where sales requests a customer exception that operations cannot support. Everyone agrees to work through the details, but no one establishes who will decide or when sales will receive an answer. A week later, the request returns. Shortening the meeting would save time without resolving the conflict.
A standing forum determines when information crosses boundaries and when a problem gets access to authority. It also determines whether a commitment reaches people who were not in the room. In that sense, the meeting is part of the operating system. Its value depends on what happens before, during and after the conversation.
Give each forum a job
Andrew S. Grove distinguishes two kinds of meetings in High Output Management: recurring, process-oriented meetings that exchange information and meetings convened to address a particular problem. A staff meeting may reveal a delivery risk early enough for others to adapt. A one-on-one may help someone understand a difficult situation. Neither has failed because it ends without a decision. A meeting called to settle conflicting customer and capacity commitments has a different job.
Calendars often conceal that difference. A leadership meeting begins with reports, reaches a capacity conflict and ends just as participants discover that the person authorized to decide is absent. If the forum’s purpose was to surface risks, the exchange may have been useful. If its purpose was to settle trade-offs, the team needs to change who attends, where the choice is made or how quickly the question moves there.
That distinction makes it easier to judge whether people need to assemble at all. A routine update may be better read than presented. A choice within one leader’s authority may require consultation rather than an executive audience. Shared time earns its place when people need to test conflicting information, understand consequences for other functions or make commitments that depend on one another. The purpose should shape the invitation and the preparation, rather than leaving both to calendar habit.
Design for what happens afterward
The proposed OutcomesLab Meeting Production System follows the chain from preparation through follow-through. It is a way to inspect a forum’s purpose, pre-work, decision rights, output and closure together. Improving the agenda alone will do little if a question arrives without the evidence or authority needed to advance it. Equally, a sound choice may change nothing if it stays with the people who made it.
For the hypothetical customer exception, participants need to know what sales wants to promise, which delivery constraint stands in the way and what alternatives each function can support. Preparation should make that trade-off visible without requiring everyone to endorse an answer in advance. The point is to give challenge something concrete to work on, so time together can be spent on the disagreement rather than reconstructing the request.
The group also needs to know who can decide. The sales leader may understand the commercial consequence of declining the request; the operations leader may understand the capacity risk. Both perspectives matter, but exchanging them does not by itself grant either leader authority over the shared trade-off. If the decision belongs elsewhere, the meeting should identify what recommendation or unresolved question will go to that person, and when. Otherwise, a useful conversation can be mistaken for permission to act.
What comes out of the room depends on what the group learns. Executives might approve an exception within a capacity limit, decline it and authorize a different offer, or find that a crucial fact is missing. In the last case, progress means naming who will obtain the fact and when an authorized person will decide. Sending sales and operations away to work it out simply returns the conflict to the two functions whose obligations are pulling in different directions.
A short record can state the choice, its limits, what remains open and who must act. Affected teams need the same account. Otherwise, sales may hear permission to offer flexibility while operations hears only a request to explore it. The meeting felt settled, but the disagreement has acquired another week of life.
AI-assisted meeting tools can make that record easier to create. Features in tools from Google and Microsoft, for example, can help draft minutes and identify possible actions from a discussion. That can be particularly helpful when the alternative is a record assembled days later from memory. An accountable person still needs to check whether a proposed action was actually agreed, whether its owner accepted it and whether any condition on the decision survived the summary. The team must also decide whether a sensitive conversation is appropriate for recording or automated processing. A useful draft reduces the work of documentation; it cannot settle a trade-off that participants left unresolved.
The record remains an intention until someone checks what happened. At the next relevant review, did the offer change, was capacity reserved or was the exception declined? If action did not follow, the team can examine the obstacle rather than repeat the original debate without explanation. At OutcomesLab, we use Alignment Debt to describe the cost of unresolved differences in interpretation and commitment. A recurring meeting can add to that debt when apparent agreement produces incompatible action. It can support Decision Velocity when a sound choice reaches the people who must carry it out and stays clear enough for them to act on it.
There is some evidence for taking follow-through seriously, though not for treating this design as a guaranteed fix. In its examination of recurring performance reviews at federal agencies, the U.S. Government Accountability Office found that identifying follow-up actions was positively associated with agencies’ reported performance-improvement impact. That finding does not show that an action log causes better performance, or that federal reviews work like an executive team’s meetings. It does give leaders a reason to ask what happens after the review.
Follow the issue across meetings
Even a well-run forum can sit inside a slow system. Suppose the customer exception passes through an operating review, an executive meeting and an investment committee. Each group may have a legitimate role, while the delay accumulates between them: the first cannot decide, the second lacks a necessary fact and the third receives a changed account of the request. Better agendas in all three rooms would not necessarily shorten that path.
Audit the meeting portfolio by tracing a few consequential issues rather than counting calendar hours alone. Where did each issue first surface? When did the relevant evidence reach someone with authority? Which forum changed the course of action, and how did affected people find out? The answers might justify replacing a report with a written update, delegating a choice or clarifying which of two forums owns a trade-off. They might also show why a meeting must stay: it is the only place where competing commitments become visible together.
There is a serious risk in specifying outputs too tightly. A demand for closure can push leaders to decide before they understand the problem. Creative exploration, coaching and crisis coordination may need a looser conversation; sometimes the useful output is a better account of what remains uncertain. Nor is adding another review automatically helpful. A theoretical analysis of coordination meetings, Too Many Meetings?, models the trade-off between resolving blockers and interrupting productive work. It does not prescribe a cadence for every team, but the tension is a useful test when proposing another standing slot.
Start with one recurring meeting where an important issue keeps returning. Decide what work the forum exists to produce, then trace where progress stalls: before the room, inside it or afterward. Change that part of the process and watch what happens over the next few cycles. If people can make compatible choices and carry them through without the meeting, retire it. If the meeting is where those choices become possible, protect that work and make its commitments travel.