# Product Management Mastery System

This is a book-scale, evidence-backed learning and interview-practice system for product management. It is designed to build judgment, not framework memorization.

Research baseline: **July 28, 2026**

Begin with [Start Here](00-start-here/START-HERE.md). It routes you to the
right roadmap, placement check, question mode, and learning ledger.

## Read it as a website

The included PM Fieldbook renders this Markdown corpus as a local, searchable
reader:

```powershell
npm install
npm run dev
```

Open `http://127.0.0.1:4174/site/`. Use `Ctrl + K` on Windows to jump between
sections and question banks. Local Markdown links open inside the reader rather
than downloading.

## What this system contains

1. **Start here** — scope, evidence rules, diagnostic, and usage instructions.
2. **Roadmaps** — sequenced plans for foundations, interview preparation, mastery, portfolio work, and role-specific specialization.
3. **Resources** — 352 annotated link entries plus the audited pre-existing
   collection, covering articles, books, courses, standards, datasets, tools,
   and case-study libraries.
4. **Frameworks** — practical answer methods, decision tools, rubrics, and anti-patterns.
5. **Question banks** — 400 validated, decision-grade prompts in each of 20
   domains, with standalone worked answers authored and independently reviewed
   in auditable batches.
6. **Case studies** — evidence-backed worked cases and unsolved practice cases across product types.
7. **Templates** — reusable artifacts for discovery, strategy, metrics, experiments, roadmaps, launches, decisions, and postmortems.
8. **Reference data** — estimation anchors, metric definitions, business-model equations, and technical primers.
9. **Progress** — an auditable manifest of planned, drafted, reviewed, and verified material.

## Committed 400-question banks

The system treats “everything I might have missed” as a real scope requirement. The corpus therefore covers twenty distinct competencies. Industry, geography, company stage, and product archetype are tags and learning paths—not excuses to repeat the same question with a different brand name.

| # | Code | Domain | Questions |
| ---: | --- | --- | ---: |
| 1 | PDS | Product sense and PM product design | 400 |
| 2 | GUE | Guesstimation and market sizing | 400 |
| 3 | RCA | Root-cause analysis and product diagnosis | 400 |
| 4 | MET | Metrics, KPI design, and product analytics | 400 |
| 5 | EXP | Experimentation and causal inference | 400 |
| 6 | DIS | Product discovery and user research | 400 |
| 7 | STR | Product strategy, vision, and competition | 400 |
| 8 | PRI | Prioritization and roadmapping | 400 |
| 9 | EXE | Execution, delivery, product operations, and lifecycle | 400 |
| 10 | GRW | Growth, activation, engagement, and retention | 400 |
| 11 | MON | Monetization, pricing, and unit economics | 400 |
| 12 | GTM | Go-to-market, launch, and product marketing | 400 |
| 13 | TEC | Technical PM, systems, APIs, and data | 400 |
| 14 | AIP | AI/ML and generative-AI product management | 400 |
| 15 | PLT | Platforms, marketplaces, and ecosystems | 400 |
| 16 | B2B | B2B and enterprise product management | 400 |
| 17 | BEH | Leadership, behavioral, and stakeholder management | 400 |
| 18 | COM | Product communication, writing, and executive influence | 400 |
| 19 | TSP | Trust, safety, privacy, accessibility, and product ethics | 400 |
| 20 | LDR | Product leadership, portfolio, and operating model | 400 |
|  |  | **Total** | **8,000** |

All twenty 400-row registers pass the corpus validator. A register row is a
question prompt, not a completed answer. Each authored answer is a standalone
Markdown file with the same stable ID; its state is separately tracked as
`drafted`, `reviewed`, or `verified`. This preserves the parent
`interview-practice/README.md` rule that one answer lives in one file and keeps
coverage from being mistaken for completion.

The [question-register specification](00-start-here/QUESTION-REGISTER-SPEC.md)
keeps all 400 prompts, context quotas, and answer states auditable while
standalone answers are authored and reviewed in batches.

## Definition of a completed question

A question is not marked complete merely because a prompt exists. A completed item includes:

- metadata: domain, subdomain, difficulty, target time, product context, geography, and skills tested;
- the prompt and its ambiguity traps;
- clarifying questions and a declared scope;
- a structured approach and rationale for choosing it;
- a fully worked answer with explicit assumptions, calculations, or evidence;
- alternatives, trade-offs, failure modes, and edge cases;
- success metrics and guardrails where relevant;
- common weak-answer patterns and how to improve them;
- interviewer follow-ups with answer guidance;
- a question-specific scoring rubric;
- sources for factual claims, with assumptions clearly separated from public facts.

## Evidence policy

Sources are tiered so that a polished but unsupported blog post does not outrank a standard or primary case:

- **Tier A:** standards, regulators, government data, peer-reviewed research, official documentation, public filings, and first-party engineering/product accounts.
- **Tier B:** original practitioner frameworks, high-quality books, credible interviews with direct participants, and well-documented case studies.
- **Tier C:** aggregators, community collections, interview-prep sites, and secondary summaries. These are useful for prompt discovery but cannot be the sole support for a factual claim.

Every source record includes publisher, URL, topic, access level, evidence tier, freshness, and a “best for” note. Time-sensitive claims carry an as-of date.

## Important corpus boundaries

- Existing Next Leap coursework remains in `../../next-leap/` and is linked, not duplicated.
- The private course-submission archive and browser-profile data are never copied into this corpus.
- Worked interview scenarios clearly distinguish **public fact**, **reasonable estimate**, and **hypothetical interview assumption**.
- Frameworks are tools, not scripts. Strong answers may depart from them when the problem requires it.

## Current status

See [STATUS.md](08-progress/STATUS.md) for the live delivery ledger and [DELIVERY-CONTRACT.md](00-start-here/DELIVERY-CONTRACT.md) for exact scope and quality gates.