The Capacity Ledger

The fracture you can’t see

The plan says green. The calendar says crowded.
Interrupts arrive through every side door: customer pings, exec drive-bys, dependency surprises.

Ad-hoc feels harmless in the moment. It compounds across a week.
Context switching taxes every estimate until timelines wobble.
Nobody is slacking. The system is uncounted.

Quinn’s move: surface reality before enforcing rules. Count what actually happens. Then change how work enters and moves.

The receipts before the rules

Quinn starts with evidence. One week. No judgment.

• Capacity receipts: each person marks hours as Planned, Interrupt, Ad-hoc.
• Calendar trace: recurring meetings and status time tagged as Operating Cost.
• Entry path scan: list how new work shows up and who can say yes.
• Change log: every mid-week change gets a one-line receipt with who/why/when.

Patterns appear fast: the interrupt band is bigger than the plan. Ad-hoc slips through because it is “small.” Status time blooms when the plan drifts.

The Capacity Ledger (how it works)

A single, living page per team. Updated weekly. Visible to all.

• Baseline load: 60 to 70% reserved for planned work. Name the hours. Do not exceed.
• Interrupt tax: fixed weekly band for interrupts. If it fills, trade-offs happen.
• Ad-hoc cap: a small, explicit ceiling. Overages roll forward, not sideways.
• Freeze windows: once the week starts, changes travel one path with one owner and a response timer.
• Commit rules: each planned item has entry criteria, estimate range, owner, and acceptance proof.
• Decision log: changes recorded with receipts, timestamps, and the decision owner.

The ledger is not a dashboard. It is an operating contract grounded in receipts.

The weekly cadence

Monday set: Fill the baseline with planned work only. Publish interrupt/ad-hoc caps.
Midweek control: All new work follows the single path. Exceptions are documented.
Friday review: Compare plan vs receipts. Tune the bands. Log deltas and decisions.

What leaders manage (not micromanage)

• Trade-offs in daylight: When the interrupt band is full, something gives. Decide once, visibly.
• Owner and path: One path in, one owner, one timer. No parallel overrides.
• Focus protection: Freeze windows keep hands steady. Exceptions are rare and recorded.
• Signal integrity: Stop reading velocity alone. Track plan reliability and overage patterns.

Evidence you can track (starter set)

• Plan reliability: planned items delivered / planned items committed.
• Interrupt share: interrupt hours / total hours.
• Ad-hoc overage: hours beyond the cap; show the delta.
• Throughput per planned hour: delivered units / planned hours.
• Receipts rate: % of individuals with complete receipts by week’s end.
• Decision time: median time from new ask → accept/deflect decision.

Implementation notes (what actually helps)

• Start small: one team, two weeks, ruthless clarity.
• Name the bands: numbers beat vibes. Tune after two cycles.
• Publish the rules: write the entry criteria and freeze windows in plain language.
• Train the path owner: authority to say “not this week” with alternatives.
• Review with receipts: show before/after charts in the same view as decisions.

What changes when it works

• Plans fit inside the week.
• Ad-hoc stops hijacking focus.
• Interrupts are priced, not ignored.
• Status meetings shrink; decision logs grow.
• Predictability climbs; blame loops fade.

Cheat sheet (print and tape up)

Baseline load. Interrupt tax. Ad-hoc cap. Freeze windows.
One path. One owner. One timer. Decision logged with receipts.

Quinn’s note to operators: Treat capacity like cash. Budget it. Track it. Defend it.

Next Episode: The Proofline Dashboard

A compact, receipts-first view of Receipts Rate, Lane Integrity, and Decision Velocity, refreshed monthly so leaders steer with evidence, not noise.