Skip to content

MODEL 01 / CAPACITY & CONSEQUENCES

The Commitment Spiral

Explore how promises, shortcuts and recovery shape the weeks that follow.

Build a custom policy sequence

Week 6 / 36

Paused
Usable output—points delivered & sound
Open promises—points still to deliver
Team capacity—of 12 points, before overtime
Trust—an illustrative index / 100

THE WEEKS ADD UP

Work waiting

Dashed lines: continuation
Read chart data

This week’s capacity

SAME DEMAND. ANOTHER COURSE.

What would change?

Freeze the current policy as a reference. Your next changes affect only your experiment.

Challenge the modelThree illustrative assumptions

What if these feedbacks are weaker—or stronger? Changes replay both courses from week 1 using the same assumptions. Your policies, seed and current week are kept.

Effort to repair each defective point. Illustrative default: 1.7 ×.

Capacity lost at maximum fatigue. At 0%, fatigue cannot reduce capacity. Default: 45%.

Extra effort per 10 overdue points. Trust-related reporting remains. Default: 0.65.

Inside the modelMechanism, assumptions & limits

The loop

Accept work → delivery pressure → overtime and shortcuts → fatigue and delayed repairs → less room for the next promise.

Promises are due by the end of the following week. Returned defects are repaired first. Late promises reduce trust; lateness and lower trust add reporting work.

Declining work has a cost too: it never becomes output. Demand is external, so every fork sees exactly the same demand.

The accounting

Capacity = 12 × (1 − 0.45 × fatigue)
Capacity + overtime = reporting + recovery + repairs + delivery effort + idle

Shortcuts reduce effort per delivered point but increase defects. Defects return two weeks later and cost 1.7 effort points each to repair. “Usable output” subtracts all outstanding defects, including hidden ones.

Reporting costs 0.4 effort points, plus 0.65 per 10 overdue points and up to 1.4 for lost trust. Total reporting is capped at 40% of capacity.

Recovery protects 0–25% of capacity. At 25%, overtime is off. Spare capacity also helps fatigue fall.

A model, not a forecast

Units, coefficients and scenarios are invented. Work is divisible and alike. No staff changes, deadlines renegotiated, strategic benefits from accepting work, or revenue model.

The full equations are readable in the engine. Results show what follows from these assumptions, not what will happen to a real team.

Your experiment is saved in this browser. Export creates a Markdown record of the policies, assumptions and outcomes.