making-product-decisions（做出产品决策）

Making Product Decisions - Structured Decision Framework

A meta-framework for making and documenting product decisions. Combines decision science principles with practical product management needs to ensure better decisions, stakeholder alignment, and organizational learning.

When to Use This Skill

• Choosing between competing priorities or approaches
• Making irreversible or high-stakes decisions
• Aligning stakeholders with different perspectives
• Documenting decisions for future reference
• Evaluating past decisions for learning
• Delegating decision-making authority

Core Concepts

Decision Types (Bezos Framework)

+------------------+------------------+
|   Type 1         |   Type 2         |
|   (One-way door) |   (Two-way door) |
+------------------+------------------+
| Irreversible     | Reversible       |
| High stakes      | Lower stakes     |
| Slow, careful    | Fast, iterate    |
| Senior decision  | Delegate widely  |
+------------------+------------------+

Decision Quality vs. Outcome

Judge decisions by process quality, not just outcomes.

Data-Informed vs. Data-Driven

Most product decisions should be data-informed, not purely data-driven.

Analysis Framework

Step 1: Frame the Decision

Step 2: Generate Options

Always have at least 3 options:

1. Do nothing / status quo
2. Option A
3. Option B

Avoid binary framing - it limits thinking.

Step 3: Establish Criteria

Step 4: Evaluate Options

Step 5: Document and Decide

Record:

• Decision made
• Rationale
• Dissenting views
• Success criteria
• Review date

Output Template

## Product Decision Record

**Decision:** [Clear statement of what was decided] **Date:** [Date] **Decision
maker:** [Name] **Status:** [Proposed/Approved/Implemented]

### Context

**Problem/Opportunity:** [What prompted this decision]

**Constraints:** [Time, resources, dependencies]

**Reversibility:** [Type 1 / Type 2]

### Options Considered

| Option     | Description | Pros | Cons |
| ---------- | ----------- | ---- | ---- |
| Status quo | [Desc]      | [+]  | [-]  |
| Option A   | [Desc]      | [+]  | [-]  |
| Option B   | [Desc]      | [+]  | [-]  |

### Decision Criteria

| Criterion | Weight | Rationale |
| --------- | ------ | --------- |
| [C1]      | [1-5]  | [Why]     |
| [C2]      | [1-5]  | [Why]     |

### Evaluation

| Option  | [C1]  | [C2]  | Weighted Total |
| ------- | ----- | ----- | -------------- |
| [Opt 1] | [x/5] | [x/5] | [Score]        |
| [Opt 2] | [x/5] | [x/5] | [Score]        |

### Decision

**Chosen option:** [Option name]

**Rationale:** [Why this option best meets criteria]

**Dissenting views:** [Captured disagreements and concerns]

### Success Criteria

| Metric | Current | Target  | Measure By |
| ------ | ------- | ------- | ---------- |
| [M1]   | [Value] | [Value] | [Date]     |

### Review

**Review date:** [Date] **What we'll evaluate:** [Criteria for success/failure]

Real-World Examples

Example 1: Build vs. Buy

Decision: Build custom analytics or use third-party tool?

Decision: Buy, despite customization limitations.

Example 2: Feature Prioritization

Decision: Next quarter focus - mobile app or API improvements?

Applied decision criteria:

• Revenue impact (weight: 5)
• User retention (weight: 4)
• Strategic positioning (weight: 3)
• Engineering complexity (weight: 2)

Result: Mobile app scored higher on revenue and retention despite higher complexity.

Best Practices

Do

• Make decision criteria explicit before evaluating
• Include "do nothing" as an option
• Document dissenting opinions
• Set review dates for major decisions
• Separate decision quality from outcome

Avoid

• Analysis paralysis on Type 2 decisions
• HiPPO (Highest Paid Person's Opinion) decisions
• Retroactive justification
• Ignoring intuition entirely
• Forgetting to review past decisions

Decision Speed Guidelines

Integration with Other Methods

Resources

• Thinking in Bets - Annie Duke
• Decisive - Chip & Dan Heath