The percentage of topics in a meeting that reached full closure — where every required atomic signal for that topic type was present and the topic could be marked achieved.
Direct Answer
Closure Reliability is the percentage of topics in a meeting that reached full closure. A topic is fully closed when it has accumulated all the atomic signals required for its closure type — and can therefore be marked achieved. Not every topic type requires the same signals. A decision, an action, an escalation, and a deferral each have different structural requirements. Closure Reliability holds each one to its own bar. The score tells you what fraction of what the meeting touched actually landed.
A topic is marked achieved when it has all required atomic signals for its type — who owns it, what happens next, when, and under what conditions. When all required signals are present, the topic can be marked achieved. When one or more are missing, it cannot, regardless of how resolved the conversation felt in the room.
The gap between those two states is where execution consistently fails. Sam Kaner, lead author of Facilitator's Guide to Participatory Decision-Making (Jossey-Bass, 2007), calls it "false agreement" — participants nod along and leave the room with entirely different interpretations of what was decided. Patrick Lencioni, in Overcoming the Five Dysfunctions of a Team (Jossey-Bass, 2005), named the same pattern the "kiss of yes": polite agreement in the room, no structural clarity to follow through on it. When a group closes a topic without spelling out implementation specifics, Kaner describes the result as a "pseudo-solution" — an illusion of closure that forces the group back to the same topic in a future meeting. Closure Reliability is built to detect exactly this gap. It does not measure whether the conversation felt resolved. It measures whether the structural conditions for resolution were actually present.
Each closure type has its own bar. A topic is only marked achieved when every required signal for its type is present. This approach mirrors what the McKinsey Team Effectiveness Index, drawn from research by Aaron De Smet and colleagues (2024), identifies as a measurable indicator of decision health: high-performing teams end meetings with an explicit summary of what was decided, who owns the next step, and the rationale behind it. Kaner is more direct about the consequence of skipping this: if a group stays vague on what needs to be done, who will do it, and by when, "no one takes responsibility, and nothing happens." The required signals for each closure type are the minimum structural specifics needed to make that outcome impossible to avoid:
Aligned
Requires an explicit recap or confirmation spoken out loud and quotable. Polite assent and silence are flagged as failures for this type — the system treats them as absence of the signal, not presence. Common in Status Updates, Planning Meetings, Sync Meetings, Retrospectives, Decision Forums, Conflict Resolution, Coaching Feedback, and Backlog Refinement sessions.
Action Taken
Requires both an owner and a next step — both signals must be present. If the team discusses an issue and concludes "we should fix that" but no one is explicitly assigned, the closure fails. The conversation happened; the coordination did not. Common in Planning Meetings, Sync Meetings, Retrospectives, Backlog Refinement, Coaching Feedback, Conflict Resolution, Project Reviews, and Technical Architecture sessions.
Decision Made
Requires the decision to be explicitly stated as a binding commitment. Discussing options or stating a preference does not satisfy this signal. The group must land on something and name it as decided. Common in Decision Forums, Planning Meetings, Project Reviews, and Technical Architecture sessions.
Parked / Deferred
Requires a next step that explicitly routes the revisit. Saying "let's park it" alone is not enough. The group must name the forum, action, or trigger — for example, "let's park this until we have the Q3 budget numbers." Without that, the topic becomes a Zombie Topic by definition. Primarily expected in Decision Forums.
Escalated
Requires both an owner and a next step — who is escalating the conflict or blocker, and exactly how they are doing it. Without both signals, the escalation has no one carrying it and no path forward. It becomes an Unrouted Escalation. Common in Decision Forums and Project Reviews.
A low score means a significant share of what the meeting touched did not reach achieved status. Those topics are still open — structurally, even if not in the minds of the people who were in the room. They are the primary source of decision churn: the same conversations resurfacing week after week because the first conversation never actually closed. Research on remote collaboration links this directly to project stagnation — without explicit ownership and deadlines established before the meeting ends, work stops at the threshold between agreement and execution.
If a team is experiencing persistent decision churn, the most reliable structural intervention is a hard stop five minutes before the end of every meeting. No adjournment until every nodded agreement has been translated into a named owner and a deadline. That single habit changes what Closure Reliability measures, which changes what it produces.
Closure Reliability is a percentage, and percentages can obscure what matters. A meeting that closes ten routine administrative topics and fails to close two high-stakes strategic decisions might score 83% — a number that looks healthy while masking a serious execution risk. The percentage is a useful aggregate signal. The topic-level breakdown is where the actual risk lives.
There is also a low-stakes exception worth noting. Kaner observes that for genuinely minor, routine issues, pushing for rigorous documented mechanics wastes the group's time. Holding a quick administrative update to the same structural bar as a cross-functional decision creates coordination drag without adding reliability. Closure Reliability is most valuable as a diagnostic for consequential topics — decisions, delegated actions, escalations, and deferrals where incomplete closure has a real downstream cost.
Closure Reliability is its own standalone metric — the straight percentage of achieved topics. The Reliability dimension inside Coordination Quality is a different calculation: it applies weighting to reflect coordination debt (full closures at full value, partial closures at a fraction). They measure related but distinct things. Closure Reliability tells you how many topics crossed the achieved threshold. The Reliability dimension inside Coordination Quality tells you the quality-adjusted closure rate, penalising closures that look complete but aren't.
Kaner, S., Lind, L., Toldi, C., Fisk, S., & Berger, D. (2007). Facilitator's guide to participatory decision-making (2nd ed.). Jossey-Bass.
Lencioni, P. (2005). Overcoming the five dysfunctions of a team: A field guide for leaders, managers, and facilitators. Jossey-Bass.
De Smet, A., D'Auria, G., Meijknecht, L., & Albaharna, M. (2024). Go, teams: When teams get healthier, the whole organization benefits. McKinsey & Company.
"If a group stays vague on what needs to be done, who will do it, and by when — no one takes responsibility, and nothing happens." — Sam Kaner
Closure Reliability is the percentage of topics in a meeting that reached full closure — meaning they had all the atomic signals required for their closure type to be marked achieved. Not all closure types require the same signals. A decision requires different specifics than an action or an escalation. Closure Reliability measures whether each topic crossed its own bar, not a universal one.
Achieved means the topic accumulated all required atomic signals for its closure type. An Aligned closure requires an explicit, quotable recap — polite assent doesn't count. An Action Taken closure requires both a named owner and a next step. A Decision Made closure requires the decision to be stated as a binding commitment, not just a preference. A Parked or Deferred closure requires a named trigger or forum for revisiting. An Escalated closure requires both a named owner and a defined path for escalation. When all required signals are present, the topic is marked achieved. When one or more are missing, it is not.
Because different closure types create different coordination obligations. An action that nobody owns will not happen. A deferral without a trigger becomes a Zombie Topic. An escalation without a named carrier becomes an Unrouted Escalation. Requiring the same signals from every closure type would either overburden simple closures or underspecify complex ones. Each type is held to the bar that matches what it needs to survive past the meeting.
Directly. Topics that were discussed but not marked achieved — because a required signal was missing — are the primary source of decision churn. They look closed in the room but lack the structural completeness to survive handoff. High Closure Reliability means most topics crossed their bar. Low Closure Reliability is a leading indicator that churn is coming.
the per-meeting composite score across all five coordination dimensions
the partial closures most likely to resurface and consume future meeting time
items parked without a next step that will return in a future meeting
topics that came back in this meeting after being parked or escalated before
What makes team decisions durable.
ResearchThe structural patterns behind decision churn.
Category DefinitionThe coordination layer and how it's measured.
Community ResearchPractitioners on closure in practice.
New articles on coordination dynamics, decision reliability, and the science of how teams actually work.
Subscribe to our newsletter