product-architecture

Product Architecture System

"A roadmap is not a feature list. It's a sequence of bets on how to create value."

This skill covers the Product System — structuring what you're building and why now. It turns validated opportunities into bets, organizes work into capability blocks, and creates roadmaps that communicate strategy without false precision.

Part of: Modern Product Operating Model — a collection of composable product skills.

Related skills: product-strategy, product-discovery, product-delivery, ai-native-product, product-leadership

When to Use This Skill

Use this skill when:

• Organizing your product into capability blocks
• Converting discovery opportunities into prioritized bets
• Building or updating your roadmap
• Writing solution briefs before engineering commits
• Preparing for planning cycles (quarterly, annually)
• Communicating product direction to stakeholders

Cadence: Quarterly planning, ongoing refinement | Owner: PM with trio input

The Problem This Solves

Most teams have:

1. Feature lists masquerading as roadmaps
2. No clear connection between strategy and what gets built
3. Solution briefs that are either too vague or too prescriptive
4. Backlogs organized by urgency, not impact

The Product Architecture System creates a clear hierarchy: Strategy → Bets → Blocks → Features, with traceable connections at each level.

Philosophy

Core Beliefs

1. Bets over features — We're not building features, we're placing bets on outcomes
2. Roadmaps show direction, not dates — Commitments are for sprints, not quarters
3. Problems before solutions — Every bet must connect to a validated opportunity
4. Solution briefs over PRDs — Just enough spec to start, emergent detail through building
5. Blocks organize complexity — Group capabilities by customer value, not technical architecture

What This Framework Rejects

• Feature factories (build what's requested, not what matters)
• Date-driven roadmaps (false precision creates false expectations)
• Disconnected backlogs (no traceability to strategy)
• Waterfall PRDs (200-page specs nobody reads)
• Tech-driven architecture (organizing by system, not customer)

Framework Components

1. Product Blocks & Features

What is a Block?

A block is a capability area that delivers distinct customer value. Blocks organize your product by what customers can accomplish, not by technical architecture.

Good Block Examples:

• "Onboarding" (helps users get started)
• "Reporting" (helps users understand performance)
• "Collaboration" (helps teams work together)
• "Integrations" (connects to existing workflows)

Bad Block Examples (Tech-Driven):

• "Backend services"
• "API layer"
• "Database module"

Block Portfolio View:

| Block | Owner | Maturity | Strategic Priority | Active Bets | |-------|-------|----------|-------------------|-------------| | Onboarding | [PM] | Growth | P1 | 2 | | Reporting | [PM] | Mature | P2 | 1 | | Collaboration | [PM] | New | P1 | 3 | | Integrations | [PM] | Growth | P3 | 0 |

Block Maturity Stages:

| Stage | Definition | Focus | |-------|------------|-------| | New | Unproven, high uncertainty | Find PMF within block | | Growth | Validated, expanding | Scale and optimize | | Mature | Stable, well-understood | Maintain, incremental improvements | | Sunset | Declining value | Deprecate or replace |

Features Within Blocks:

Each block contains features. Features are the specific capabilities users interact with.

BLOCK: Reporting
├── Feature: Dashboard builder
├── Feature: Scheduled reports
├── Feature: Export to PDF
└── Feature: Custom metrics

0→1 Mode: One block, maybe two. Don't over-structure before you have PMF.

Scaling Mode: Multiple blocks with clear owners. Block health reviews quarterly.

2. The Bet Backlog

What is a Bet?

A bet is a time-boxed investment with a hypothesis, success metrics, and kill criteria. Unlike features, bets explicitly acknowledge uncertainty.

Bet Format:

BET: [Name]
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
Hypothesis: We believe that [action] will result in [outcome]
Target metric: [Metric] from [baseline] to [target]
Timebox: [Duration]
Block: [Which block this belongs to]
Opportunity: [Link to validated opportunity in OST]
Kill criteria: [When we stop]
Scale criteria: [When we double down]
Effort: [T-shirt size]
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━

Bet Categories:

| Category | Description | Example | |----------|-------------|---------| | Value creation | New capabilities that solve customer problems | New reporting feature | | Growth | Acquisition, activation, expansion | Onboarding improvements | | Platform | Infrastructure, scalability, efficiency | Performance optimization | | Trust | Security, compliance, reliability | SOC 2 certification | | Moat | Building defensibility | Proprietary data features |

Portfolio Balance:

| Category | Target Allocation | Rationale | |----------|-------------------|-----------| | Core (proven, incremental) | 70% | Keep the lights on, serve existing customers | | Adjacent (related, moderate risk) | 20% | Expand to new use cases or segments | | Transformational (new, high risk) | 10% | Explore future opportunities |

Bet Prioritization:

Use ICE, RICE, or simple compare-and-contrast:

| Bet | Impact | Confidence | Effort | Score | Priority | |-----|--------|------------|--------|-------|----------| | A | High | High | Medium | [X] | P0 | | B | High | Low | Low | [X] | P1 | | C | Medium | High | Low | [X] | P1 |

Backlog Rules:

• Every bet traces to a validated opportunity
• Maximum 3 P0 bets at any time
• Bets without kill criteria don't make the backlog
• Review and reprioritize quarterly (or when evidence changes)

3. Roadmap

Roadmap Philosophy:

A roadmap is a communication tool, not a contract. It shows direction and priorities, not delivery dates.

Roadmap Formats:

| Format | Audience | Purpose | |--------|----------|---------| | Now / Next / Later | Team, stakeholders | Current priorities without dates | | Quarterly themes | Executives, board | Strategic direction by time horizon | | Outcome roadmap | Team, stakeholders | What outcomes we're targeting when |

Now / Next / Later Format:

NOW (Current quarter - high confidence)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
• [Bet A] — [Brief description]
• [Bet B] — [Brief description]

NEXT (Next quarter - medium confidence)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
• [Bet C] — [Brief description]
• [Bet D] — [Brief description]

LATER (Future - low confidence, subject to change)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
• [Bet E] — [Brief description]
• [Bet F] — [Brief description]

Quarterly Themes Format:

| Quarter | Theme | Key Bets | Target Outcome | |---------|-------|----------|----------------| | Q1 | "Foundation" | [Bets] | [Outcome metric] | | Q2 | "Scale" | [Bets] | [Outcome metric] | | Q3 | "Expand" | [Bets] | [Outcome metric] | | Q4 | "Optimize" | [Bets] | [Outcome metric] |

Outcome Roadmap Format:

Instead of showing features, show outcomes:

| Timeframe | Outcome | How We'll Know | Key Bets | |-----------|---------|----------------|----------| | Q1 | Improve activation rate | 30% → 45% | [Bets focused on this] | | Q2 | Reduce churn | 5% → 3% | [Bets focused on this] | | H2 | Enter new segment | 10 customers | [Bets focused on this] |

Roadmap Anti-Patterns:

| Anti-Pattern | Why It Fails | Instead | |--------------|--------------|---------| | Date commitments | Creates false expectations | Time horizons (Now/Next/Later) | | Feature lists | No strategic context | Outcome-focused themes | | Too detailed | Can't see forest for trees | High-level, drill down on request | | Never updated | Becomes fiction | Review quarterly minimum | | No tr