Finishing is a different skill from building

And it’s the one almost everyone is worse at.

Twelve weeks, twelve domains, one argument

Software. Infrastructure. Research. Hospitality. Law. Manufacturing. Institutional handovers. Startups. Construction. Aviation. Open-source overbuilding. And ourselves.

Different mechanisms. Same shape: the closer something gets to done, the more the cost of the remaining distance stops matching the distance itself.

Weeks 1–3

  • Week 1 — Software — shipped demos that were never built to survive contact with a real user
  • Week 2 — Infrastructure — the last customer on the line costs as much to connect as every other customer combined
  • Week 3 — Graduate research — the dissertation is the one deliverable nobody else is waiting on, so it’s the easiest to keep deferring

Weeks 4–6

  • Week 4 — Hospitality — the bathroom earns a restaurant zero new customers, which is exactly why it’s the first thing to get shortchanged
  • Week 5 — Law — a case won on every argument, destroyed completely by one missed date on a calendar nobody was watching
  • Week 6 — Manufacturing — the same defect costs roughly a hundred times more once the customer finds it, and it’s a measured number, not a vibe

Weeks 7–9

  • Week 7 — Institutional handovers — two teams each finish their half correctly, and the failure lives in the seam neither team’s own review could see
  • Week 8 — Startups — the finish line relocates exactly as fast as you close the distance to it
  • Week 9 — Construction — occupancy is a legal switch, not a percentage; 98% done buys a building nothing

Weeks 10–12

  • Week 10 — Aviation — the last five percent of a flight’s duration carries roughly half its historical fatal risk
  • Week 11 — Overbuilding — past a point, continued finishing stops adding value and starts subtracting it
  • Week 12 — Synthesis — eleven mechanisms, one shared blind spot

How the assessment holds you to this

  • Labs — 20% — twelve weekly chances to practise finishing something small, not just building it
  • Assignment 1 — 30%, week 6 — find something real and unfinished, and name precisely why its last mile is expensive
  • Final project — 50%, week 12 — go back and actually cross it