making-product-decisions
Compare original and translation side by side
🇺🇸
Original
English🇨🇳
Translation
ChineseMaking Product Decisions - Structured Decision Framework
产品决策制定 - 结构化决策框架
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.
这是一个用于制定和记录产品决策的元框架。它结合了决策科学原理与实用的产品管理需求，以确保做出更优决策、对齐利益相关者并实现组织学习。
When to Use This Skill
何时使用该方法
- 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
- 在相互竞争的优先级或方案中做选择时
- 制定不可逆转或高风险决策时
- 协调持不同观点的利益相关者时
- 记录决策以供未来参考时
- 评估过往决策以从中学习时
- 下放决策权限时
Core Concepts
核心概念
Decision Types (Bezos Framework)
决策类型（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 |
+------------------+------------------++------------------+------------------+
| Type 1 | Type 2 |
| (单向门) | (双向门) |
+------------------+------------------+
| 不可逆转 | 可逆转 |
| 高风险 | 低风险 |
| 缓慢、谨慎制定 | 快速制定、迭代 |
| 高层决策 | 广泛授权 |
+------------------+------------------+Decision Quality vs. Outcome
决策质量 vs. 结果
| Good Outcome | Bad Outcome | |
|---|---|---|
| Good Decision | Deserved success | Bad luck |
| Bad Decision | Good luck | Deserved failure |
Judge decisions by process quality, not just outcomes.
| 良好结果 | 不良结果 | |
|---|---|---|
| 优质决策 | 应得的成功 | 运气不佳 |
| 劣质决策 | 运气好 | 应得的失败 |
判断决策应依据过程质量，而非仅看结果。
Data-Informed vs. Data-Driven
数据参考 vs. 数据驱动
| Approach | When to Use |
|---|---|
| Data-driven | Clear metrics, sufficient data, understood system |
| Data-informed | Incomplete data, novel situations, judgment needed |
| Intuition-led | Time pressure, expert domain, pattern matching |
Most product decisions should be data-informed, not purely data-driven.
| 方法 | 适用场景 |
|---|---|
| 数据驱动 | 指标清晰、数据充足、系统逻辑明确时 |
| 数据参考 | 数据不完整、场景新颖、需要主观判断时 |
| 直觉主导 | 时间紧迫、具备领域专业知识、可进行模式匹配时 |
大多数产品决策应基于数据参考，而非纯粹的数据驱动。
Analysis Framework
分析框架
Step 1: Frame the Decision
步骤1：明确决策边界
| Element | Question |
|---|---|
| What | What exactly are we deciding? |
| Why | Why does this decision matter? |
| Who | Who should be involved? |
| When | When must we decide by? |
| Reversibility | Type 1 or Type 2 door? |
| 要素 | 对应问题 |
|---|---|
| 决策内容 | 我们具体要决定什么？ |
| 决策原因 | 这个决策为何重要？ |
| 参与人员 | 哪些人应该参与决策？ |
| 决策时限 | 我们必须在何时做出决策？ |
| 可逆性 | 属于Type 1还是Type 2决策？ |
Step 2: Generate Options
步骤2：生成备选方案
Always have at least 3 options:
- Do nothing / status quo
- Option A
- Option B
Avoid binary framing - it limits thinking.
始终至少准备3个选项：
- 维持现状/不采取行动
- 方案A
- 方案B
避免二元对立的框架——这会限制思考。
Step 3: Establish Criteria
步骤3：确立评估标准
| Criterion | Weight | Why It Matters |
|---|---|---|
| [Criterion 1] | [1-5] | [Explanation] |
| [Criterion 2] | [1-5] | [Explanation] |
| [Criterion 3] | [1-5] | [Explanation] |
| 评估标准 | 权重 | 重要性说明 |
|---|---|---|
| [标准1] | [1-5] | [解释内容] |
| [标准2] | [1-5] | [解释内容] |
| [标准3] | [1-5] | [解释内容] |
Step 4: Evaluate Options
步骤4：评估备选方案
| Option | Criterion 1 | Criterion 2 | Criterion 3 | Total |
|---|---|---|---|---|
| Status quo | [Score] | [Score] | [Score] | [Sum] |
| Option A | [Score] | [Score] | [Score] | [Sum] |
| Option B | [Score] | [Score] | [Score] | [Sum] |
| 选项 | 标准1得分 | 标准2得分 | 标准3得分 | 总分 |
|---|---|---|---|---|
| 维持现状 | [分数] | [分数] | [分数] | [总和] |
| 方案A | [分数] | [分数] | [分数] | [总和] |
| 方案B | [分数] | [分数] | [分数] | [总和] |
Step 5: Document and Decide
步骤5：记录并确定决策
Record:
- Decision made
- Rationale
- Dissenting views
- Success criteria
- Review date
需记录：
- 最终决策内容
- 决策依据
- 不同意见
- 成功标准
- 复盘日期
Output Template
输出模板
markdown
undefinedmarkdown
undefinedProduct Decision Record
产品决策记录
Decision: [Clear statement of what was decided] Date: [Date] Decision
maker: [Name] Status: [Proposed/Approved/Implemented]
决策内容: [明确的决策陈述] 日期: [日期] 决策人: [姓名] 状态: [提议/已批准/已实施]
Context
背景
Problem/Opportunity: [What prompted this decision]
Constraints: [Time, resources, dependencies]
Reversibility: [Type 1 / Type 2]
问题/机遇: [引发此次决策的原因]
约束条件: [时间、资源、依赖关系]
可逆性: [Type 1 / Type 2]
Options Considered
考虑的备选方案
| Option | Description | Pros | Cons |
|---|---|---|---|
| Status quo | [Desc] | [+] | [-] |
| Option A | [Desc] | [+] | [-] |
| Option B | [Desc] | [+] | [-] |
| 选项 | 描述 | 优势 | 劣势 |
|---|---|---|---|
| 维持现状 | [描述] | [+] | [-] |
| 方案A | [描述] | [+] | [-] |
| 方案B | [描述] | [+] | [-] |
Decision Criteria
决策标准
| Criterion | Weight | Rationale |
|---|---|---|
| [C1] | [1-5] | [Why] |
| [C2] | [1-5] | [Why] |
| 标准 | 权重 | 依据 |
|---|---|---|
| [标准1] | [1-5] | [原因] |
| [标准2] | [1-5] | [原因] |
Evaluation
评估结果
| Option | [C1] | [C2] | Weighted Total |
|---|---|---|---|
| [Opt 1] | [x/5] | [x/5] | [Score] |
| [Opt 2] | [x/5] | [x/5] | [Score] |
| 选项 | [标准1]得分 | [标准2]得分 | 加权总分 |
|---|---|---|---|
| [选项1] | [x/5] | [x/5] | [分数] |
| [选项2] | [x/5] | [x/5] | [分数] |
Decision
最终决策
Chosen option: [Option name]
Rationale: [Why this option best meets criteria]
Dissenting views: [Captured disagreements and concerns]
选定方案: [方案名称]
决策依据: [为何该方案最符合评估标准]
不同意见: [记录的反对观点与顾虑]
Success Criteria
成功标准
| Metric | Current | Target | Measure By |
|---|---|---|---|
| [M1] | [Value] | [Value] | [Date] |
| 指标 | 当前值 | 目标值 | 评估时间 |
|---|---|---|---|
| [指标1] | [数值] | [数值] | [日期] |
Review
复盘
Review date: [Date] What we'll evaluate: [Criteria for success/failure]
undefined复盘日期: [日期] 评估内容: [判断成功/失败的标准]
undefinedReal-World Examples
实际案例
Example 1: Build vs. Buy
案例1：自研 vs. 采购
Decision: Build custom analytics or use third-party tool?
| Criterion | Weight | Build | Buy |
|---|---|---|---|
| Time to market | 5 | 2 | 5 |
| Customization | 3 | 5 | 2 |
| Long-term cost | 4 | 3 | 4 |
| Maintenance burden | 4 | 2 | 5 |
| Total | 42 | 66 |
Decision: Buy, despite customization limitations.
决策: 是自研定制分析工具还是使用第三方工具？
| 评估标准 | 权重 | 自研 | 采购 |
|---|---|---|---|
| 上市时间 | 5 | 2 | 5 |
| 定制化程度 | 3 | 5 | 2 |
| 长期成本 | 4 | 3 | 4 |
| 维护负担 | 4 | 2 | 5 |
| 总分 | 42 | 66 |
决策结果：选择采购，尽管定制化程度有限。
Example 2: Feature Prioritization
案例2：功能优先级排序
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.
决策: 下一季度重点是移动端应用还是API优化？
采用的决策标准：
- 营收影响（权重：5）
- 用户留存（权重：4）
- 战略定位（权重：3）
- 开发复杂度（权重：2）
结果：尽管开发复杂度更高，移动端应用在营收和留存维度的得分更高，因此被选定为重点。
Best Practices
最佳实践
Do
建议做法
- 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
- 在评估前明确决策标准
- 将“不采取行动”纳入备选方案
- 记录不同意见
- 为重大决策设置复盘日期
- 区分决策质量与结果
Avoid
避免事项
- Analysis paralysis on Type 2 decisions
- HiPPO (Highest Paid Person's Opinion) decisions
- Retroactive justification
- Ignoring intuition entirely
- Forgetting to review past decisions
- 在Type 2决策上过度分析导致决策瘫痪
- 依赖HiPPO（最高薪人员意见）做决策
- 事后为决策找理由
- 完全忽视直觉
- 忘记复盘过往决策
Decision Speed Guidelines
决策速度指南
| Type | Approach |
|---|---|
| Type 1, high stakes | Take time, involve stakeholders |
| Type 2, reversible | Decide quickly, iterate |
| Unclear type | Default to faster, can always slow down |
| 决策类型 | 应对方式 |
|---|---|
| Type 1、高风险 | 充足准备，邀请利益相关者参与 |
| Type 2、可逆转 | 快速决策，持续迭代 |
| 类型不明确 | 默认快速决策，必要时再放缓节奏 |
Integration with Other Methods
与其他方法的结合
| Method | Combined Use |
|---|---|
| Hypothesis Tree | Structure analysis of options |
| Jobs-to-be-Done | Ground criteria in user needs |
| Five Whys | Understand decision root causes |
| 方法 | 结合场景 |
|---|---|
| 假设树 | 结构化分析备选方案 |
| Jobs-to-be-Done | 基于用户需求确立评估标准 |
| 五问法 | 理解决策的根本原因 |