Appendix
Quick Reference: P-Cubed
Prepare → Prove → Produce
Identify What You Have
| Level | Size | Your Job |
|---|---|---|
| Problem | Undefined | Discovery → Define Initiative |
| Initiative | Weeks/Months | Decompose → Define Epics |
| Epic | Days/Weeks | P-Cubed → Produce Tasks |
| Task | 1-2 Days | Execute → Ship |
The Phases
Prepare
Discovery:
- Ask questions
- Track unknowns explicitly
- Talk to people with context
Exit when: Can explain it clearly, no more surprises, questions resolved.
Surface tradeoffs:
- What does each option optimize for?
- What are we giving up?
- What's the cost to change later?
Planning:
- Break into Tasks (1-2 days each)
- Write: "This is done when ___."
- Sequence architectural work first
Prove
Required for architectural decisions. Optional for design.
- Spike the riskiest assumption
- Validate tradeoffs in practice
- Refine the plan
Exit when: Major risks tested, approach survives scrutiny, ready to commit.
Produce
For each Task:
- Build: do the work
- Test: verify against criteria
- Ship: complete, communicate, move on
After each Task:
- What did you learn?
- Adjust remaining plan if needed
The Checklist
Before you start:
- [ ] What level is this?
- [ ] What does done look like? (Written down)
- [ ] What's load-bearing?
Before you build:
- [ ] Tradeoffs documented
- [ ] Architectural risks spiked
- [ ] Tasks scoped to 1-2 days
Before you ship:
- [ ] Criteria met (check the definition, not your feelings)
- [ ] Communicated to stakeholders
The Mantra
Define done before you start. Check criteria, not feelings. Ship when done means done.