A 2025 study published in Business Horizons identified what its authors called the meeting load paradox.

As the volume of mandatory meetings rises, participation and creative performance initially increase before declining once the cognitive and time costs exceed individual capacity. That research arrived as organizations were managing a significant expansion in scheduled calls, with meeting load peaking at approximately 21.5 hours per week in 2021 before retreating to around 14.8 hours on average, according to data tracked by Flowtrace.

A prior analysis in this series applied Amdahl's Law to that pattern, showing that compulsory synchronization functions as a serial fraction that limits organizational throughput regardless of how much parallel capacity is added through hiring or tooling.

The meeting load paradox addresses aggregate volume. A distinct but related concern involves session duration: calls that run substantially longer than their functional scope requires, particularly those extending three to six hours, carry a specific cost profile that volume-based analysis does not fully capture.

Three bodies of research converge on this pattern: the behavioral dynamics that cause duration to expand in the absence of constraints, the progressive degradation of decision quality during prolonged sessions, and the compounding recovery costs that extended calls impose on subsequent work.

Each of these operates independently; together, they produce a total cost per session that scales with duration in ways the calendar does not reflect.

Extended Call Duration as Logistical Debt


  • Meetings that lack defined objectives or completion criteria expand to fill available time, a dynamic documented since 1955 and reflected in data showing most scheduled sessions carry no stated agenda.
  • Decision quality degrades measurably during prolonged sessions as cognitive resources deplete, producing riskier choices, increased deferrals, and reduced error detection in the later portions of extended calls.
  • Meeting hangovers and post-session recovery time extend the effective cost of participation beyond the call's scheduled duration, inflating the serial fraction each attendee carries into subsequent work.
  • The cumulative serial fraction imposed by multi-hour calls, when recovery time is included, can approach or exceed half the working week at high frequency, a ceiling no staffing increase can raise.
  • Collaborators who treat scheduled duration as a measurement instrument and those who treat it as a convention will interpret the same overruns as fundamentally different kinds of events.
  • Codifying what a scheduled duration represents, whether as preference or operational metric, is the prerequisite for addressing call length as a structural problem.

Expansion Without Constraint


The baseline mechanism for unstructured meeting duration was described by British historian C. Northcote Parkinson in a 1955 essay in The Economist. This was later expanded into his 1958 book Parkinson's Law: The Pursuit of Progress.

Parkinson's central observation drew on his experience in British civil service bureaucracies. He found that organizations generate internal processes and coordination tasks that consume available capacity regardless of actual workload.

The mechanism he described is not the product of deliberate choice; it emerges from the absence of a binding completion criterion. Applied to meetings, a session scheduled for two hours to address an agenda that could be resolved in forty-five minutes will, absent active management, tend to occupy the full allocation.

"Work expands so as to fill the time available for its completion."

Data on agenda practices reflects how consistently the basic structural constraint against that expansion is absent. Research compiled by Flowtrace indicates that 64 percent of recurring meetings and 60 percent of one-off meetings are scheduled without a stated agenda.

Without defined objectives, a meeting has no internal standard against which to measure completion. Discussions extend until participants exhaust their capacity or external calendar pressure forces a stop. For calls with no fixed endpoint at all, that pressure never arrives.

Executive-level sessions concentrate this dynamic at higher per-person cost. Research by Harvard Business Review, drawn from a study tracking 27 chief executives, found that senior leaders spend an average of nearly 23 hours per week in meetings.

This is more than triple the roughly 10 hours reported in the 1960s. When a single session runs three to six hours, that event can represent a substantial share of one executive's weekly meeting allocation. It simultaneously requires equivalent attendance from others whose own schedules carry the same cost.

The downstream organizational footprint of extended executive sessions reaches well beyond the participants directly involved. The same Harvard Business Review analysis found that a single weekly senior-level meeting at one large organization consumed 300,000 person-hours per year once all preparatory and follow-up sessions were accounted for.

That figure reflects the full organizational cost of one recurring meeting, not merely the hours it placed on individual calendars. For organizations running multi-hour executive calls several times per week, the aggregate cost across all downstream sessions compounds accordingly.

The compounding effect is particularly pronounced when long calls recur on a fixed schedule. Each recurrence resets the Parkinson dynamic, because the prior session consumed its full window regardless of agenda completion.

Participants arrive with no basis to expect a shorter outcome. The absence of a tracked relationship between agenda scope and session duration means there is no signal available to either party that a shorter window would suffice.

The session length becomes self-reinforcing, and the time allocation expands with organizational tenure rather than contracting as processes mature.

More Business Articles

Output Degradation During Extended Sessions


The concern with prolonged calls extends beyond their claim on the schedule. Research on decision fatigue establishes that the quality of judgment exercised within a session declines as duration increases, in predictable and measurable ways.

Social psychologist Roy F. Baumeister and colleagues described the underlying mechanism in a 1998 paper in the Journal of Personality and Social Psychology. They found that the capacity for deliberate decision-making and self-regulation functions as a depletable resource. The more frequently that capacity is drawn upon within a session, the less of it remains for subsequent choices within the same session.

A 2025 systematic review published in Health Psychology Review identified seven cognitive-behavioral consequences that emerge as prolonged decision-making sessions continue: reduced rational thinking, increased impulsivity, riskier choices, reduced attention, reduced ability to detect errors, impaired working memory, and slowed cognitive processing.

These effects were documented primarily in clinical and high-stakes professional settings, where the consequences of degraded judgment are visible and measurable. The review's authors note that the underlying depletion mechanisms are not domain-specific, and that the pattern holds across contexts where sustained deliberation is required.

Executive sessions concentrate these effects because the decisions made within them tend to be high-stakes and interdependent. A four-hour call does not produce four hours of executive-quality deliberation.

The research on ego depletion indicates that the cognitive resources required for careful judgment are substantially exhausted well before a session of that length concludes. The individual experiencing that depletion is typically the last to recognize it.

The behavioral consequences of depletion are directionally consistent across the literature. As fatigue accumulates, individuals shift toward default or low-resistance choices, defer complex decisions to future sessions, or accept conclusions that require less cognitive effort to reach.

The later portions of a long call are more likely to produce decisions that either replicate existing patterns or require subsequent revision. An extended call that generates decisions needing follow-up sessions to revisit or correct has not accelerated the work; it has deferred the same work while adding fatigue to the next session.

Costs That Outlast the Call


The costs of a prolonged call continue after the session ends. A 2025 analysis published in Harvard Business Review identified a pattern the authors termed meeting hangovers.

These are persistent reductions in engagement and productivity that follow meetings, independent of how productive those meetings were in themselves. Survey data drawn from more than 5,000 knowledge workers found that more than 90 percent experience these effects at least occasionally, with recovery taking nearly two hours on average before focus and engagement return to baseline.

The cognitive mechanism behind that recovery lag was quantified in a 2008 study by Gloria Mark, Daniela Gudith, and Ulrich Klöcke, presented in the CHI proceedings. Research from the UC Irvine Department of Informatics summarizing that work found that it takes an average of more than 23 minutes to fully regain focused task engagement after an interruption.

The study also found that individuals compensate by increasing their pace, which maintains observable output while raising reported levels of stress, frustration, and cognitive load. This masks the real capacity loss from external view.

Research published in the Journal of Occupational and Environmental Medicine and available through PubMed Central examined the transition period between virtual meetings and subsequent work.

Allen, Thiese, Eden, and Knowles found that lower meeting satisfaction and effectiveness are associated with greater recovery time following a session. They recommended scheduling explicit gaps between calls to reduce burnout and preserve downstream productivity.

For a session extending three hours or more, the transition period scales accordingly, compounded by the cognitive depletion that accumulates throughout the call itself.

From the perspective introduced in the prior Amdahl's Law analysis on Beige Media, the serial fraction imposed by a prolonged call is substantially larger than its scheduled duration.

A three-hour session followed by a two-hour recovery period consumes five effective hours of capacity per attendee. Multiplied across the number of participants in a senior-level call and repeated several times per week, the practical serial fraction can approach or exceed half the working week. This is a ceiling on throughput that additional staff or tooling cannot raise.

Temporal Orientation and the Scheduling Gap


A further dimension of the problem arises when collaborators scheduling and attending extended calls operate from different assumptions about what a time constraint represents.

Anthropologist Edward T. Hall described two contrasting orientations toward time in his 1983 book The Dance of Life: The Other Dimension of Time. Monochronic time orientation treats schedules as binding operational structures, with adherence understood as a form of discipline and departure as a signal of misalignment.

Polychronic orientation treats tasks and conversations as having natural completion arcs that should not be arbitrarily truncated by a predefined window. Both orientations are functional in different contexts, and the distinction between them is descriptive rather than evaluative.

The operational divergence between these orientations becomes significant when schedules function as measurement instruments. A defined meeting duration is not only a logistical allocation; it encodes an implicit hypothesis about how long the task should take.

When a session consistently runs beyond its scheduled window, that overrun carries diagnostic information. The agenda scope may exceed what can be resolved in the allocated time, the decision-making process may lack sufficient structure, or the stated objectives may not require synchronous participation.

These are distinct problems with distinct remedies. They become visible as problems, however, only to someone who interprets the schedule as a measurement device in the first place.

In organizations where collaborators hold different temporal orientations, this asymmetry produces a specific class of coordination problem. One party reads duration overruns as operational data indicating structural misalignment. The other experiences the same session as a normal completion of what the call needed to accomplish.

Neither response is internally inconsistent. The disagreement about whether the call ran too long reflects a prior difference in what a schedule is understood to represent, and that prior difference is rarely made explicit.

This asymmetry is compounded by a well-documented pattern in self-assessment at the individual level. Research by organizational psychologist Tasha Eurich, published in Harvard Business Review, found that 95 percent of individuals rate themselves as self-aware.

Rigorous testing indicates the proportion who meet measurable criteria is closer to 10 to 15 percent. Eurich further found that professional seniority tends not to improve this ratio, because feedback becomes less candid as a person's standing increases.

In a peer relationship between co-founders or co-equals, the authority dynamic that suppresses candid feedback is absent, but the asymmetry in operational framework persists independently of it.

The result is a feedback environment where the costs of extended calls are clearly observable to one party and structurally obscured from the other. This is not through a hierarchy that filters information upward, but through two parties applying different interpretive frameworks to the same observable facts.

Eurich's broader finding applies here as well: blind spots are sustained in environments where feedback is informal, urgency displaces reflection, and experience is treated as a reliable proxy for accuracy. Those conditions do not require a reporting relationship to persist, and they do not resolve without an explicit shared framework for what the schedule is measuring.

The logistical debt framing extends the Amdahl framework in a specific direction. The principal of the debt is the difference between the time a session requires to accomplish its stated objectives and the time it actually consumes.

The interest is paid through degraded decision quality in the later portions of the call, recovery time that extends the effective cost per attendee, and the cumulative serial fraction that each extended session imposes on subsequent work. That interest accrues regardless of whether the participants recognize the underlying mechanism, and it compounds across every recurrence on the calendar.

The measurement gap between collaborators with different temporal orientations does not resolve through preference adjustment alone. It requires an explicit shared understanding of what a schedule represents.

Specifically, organizations must decide whether adherence to a defined duration constitutes a logistical default or an operational signal. Organizations that have codified this distinction tend to do so through structured pre-meeting documentation, explicitly scoped agendas, and post-session review of whether stated objectives were achieved within the planned window.

The organizations that have not codified it carry the debt continuously, and the absence of a line item assigned to it does not make the liability any smaller.

Sources


Article Credits