Skip to content

Coordination-Overhead Dashboard

Coordination monitor — instantiates Organization–Artifact Topology Alignment

Tracks the running cost of coordinating across boundaries — cross-team handoffs, review latency, meeting load — so misalignment shows up as a rising number before it shows up as missed releases.

Misalignment between an organization and its artifact rarely announces itself; it leaks out as a slow rise in coordination cost. Coordination-Overhead Dashboard is the standing instrument that watches for it: a small set of proxy metrics for cross-boundary friction, trended over time, wired to a review cadence that fires when a metric drifts. Its defining move is to be continuous and cost-focused — unlike the one-shot maps that describe the topology, it measures the tax the current topology is charging, and it raises an alarm early rather than diagnosing the cause.

Example

A large e-commerce org can't say whether its last reorganization helped, because "coordination pain" only ever surfaces as anecdote. It stands up a Coordination-Overhead Dashboard: cross-team pull-request review latency, the share of tickets that bounce across team boundaries, cross-team meeting hours per engineer, and lead time for changes that touch more than one service. Three months in, one area's cross-boundary ticket share creeps from roughly 15% to nearly 28% — a quiet signal that a boundary is being crossed constantly and shouldn't be. The dashboard's threshold trips a scheduled drift review of that area before it misses its quarter, when the fix is still cheap. The dashboard didn't say why; it said look here, now.

How it works

What distinguishes it from a generic engineering-metrics board is that every metric is a proxy for cross-boundary friction, and the board is coupled to a cadence:

  • Instrument a few overhead proxies — handoff latency, cross-boundary rework, meeting load — chosen because they rise when work crosses a seam it shouldn't.
  • Trend them against a baseline, so drift is visible as a slope, not a single reading.
  • Trip a drift review on threshold crossing — the second half of the mechanism is the cadence it drives, not just the chart.
  • Stop at the alarm — it flags that cost is rising; it hands off the why to the maps and the overlay.

Tuning parameters

  • Metric set — latency versus volume versus meeting-load versus rework. Each captures a different face of coordination cost; a single metric is easy to game.
  • Boundary definition — which team-and-module boundaries count as "cross." Wrong boundaries make every metric meaningless.
  • Threshold and cadence — how far a metric drifts before a review fires, and the baseline review frequency. Tight settings warn early but cry wolf; loose settings are quiet but late.
  • Normalization — per-engineer versus absolute. Absolute counts rise with headcount and mislead.
  • Leading versus lagging weighting — review-latency (leading) versus missed-deadline (lagging) signals; the more leading, the earlier the warning.

When it helps, and when it misleads

Its strength is turning a vague "we spend all day in sync meetings" into a trended number with an alarm attached, catching drift while it is still cheap to correct and giving a reorganization an actual before-and-after signal.

Its failure modes are the ones every metric board carries, sharpened here. The numbers are proxies: a low reading can mean teams have stopped coordinating in ways they should, not that alignment improved. And any metric that becomes a target gets gamed — fewer logged meetings, not less need to coordinate — which is Goodhart's law in plain view. The classic misuse is managing the dashboard instead of the coordination, or waving a green board to declare a reorganization a success. The discipline that keeps it honest is to treat the metrics as signals rather than targets, always pair a threshold crossing with the qualitative drift review it triggers, and diagnose the cause with the topology maps rather than the dashboard itself.[1]

How it implements the components

Coordination-Overhead Dashboard fills the measurement-and-cadence side of the archetype — it monitors, it does not map or diagnose:

  • coordination_overhead_metric — the trended proxy metrics it computes for cross-boundary friction.
  • drift_review_cadence — the threshold-triggered review schedule it drives, turning a rising number into a scheduled look.

It measures cost, not structure: the collective_communication_topology_map and artifact_topology_map come from Communication Pattern Review and Architecture Dependency Graph Review, and cataloguing the specific mismatches in a sociotechnical_mismatch_register is the Silo Imprint Audit's job.

References

[1] Brooks's law (Fred Brooks, The Mythical Man-Month) traces the cost of adding people to the number of communication paths, which grows on the order of n² with team size; the dashboard is watching for that coordination-cost signature. The metric is a signal, not the objective — the moment it becomes the objective, Goodhart's law degrades it. ↩