Back to Details

recommendation-canvas

Compare original and translation side by side

🇺🇸

Original

English
🇨🇳

Translation

Chinese

Purpose

目的

Evaluate and propose AI product solutions using a structured canvas that assesses business outcomes, customer outcomes, problem framing, solution hypotheses, positioning, risks, and value justification. Use this to build a comprehensive, defensible recommendation for stakeholders and decision-makers—especially when proposing AI-powered features or products that carry higher uncertainty and risk.
This is not a feature spec—it's a strategic proposal that articulates why this AI solution is worth building, what assumptions need validating, and how you'll measure success.
使用结构化画布评估并提出AI产品解决方案，该画布涵盖业务成果、客户成果、问题界定、解决方案假设、定位、风险以及价值合理性评估。利用此画布为利益相关者和决策者构建全面、具有说服力的建议——尤其是在提出具有较高不确定性和风险的AI驱动功能或产品时。
这不是一份功能规格说明书——而是一份战略提案，阐明为何值得构建该AI解决方案、哪些假设需要验证，以及如何衡量成功。

Key Concepts

核心概念

The Recommendation Canvas Framework

推荐画布框架

Created for Dean Peters' Productside "AI Innovation for Product Managers" class, the canvas synthesizes multiple PM frameworks into one strategic view:
Core Components:
  1. Business Outcome: What's in it for the business?
  2. Product Outcome: What's in it for the customer?
  3. Problem Statement: Persona-centric problem framing
  4. Solution Hypothesis: If/then hypothesis with experiments
  5. Positioning Statement: Value prop and differentiation
  6. Assumptions & Unknowns: What could invalidate this?
  7. PESTEL Risks: Political, Economic, Social, Technological, Environmental, Legal
  8. Value Justification: Why this is worth doing
  9. Success Metrics: SMART metrics to measure impact
  10. What's Next: Strategic next steps
该画布是为Dean Peters的Productside课程“产品经理的AI创新”所创建，它将多个PM框架整合为一个战略视图：
核心组件：
  1. 业务成果： 对业务有何价值？
  2. 产品成果： 对客户有何价值？
  3. 问题陈述： 以用户角色为中心的问题界定
  4. 解决方案假设： 带有实验的如果/那么假设
  5. 定位陈述： 价值主张与差异化
  6. 假设与未知因素： 哪些因素可能使方案失效？
  7. PESTEL风险： 政治、经济、社会、技术、环境、法律层面
  8. 价值合理性： 为何值得实施此方案
  9. 成功指标： 衡量影响的SMART指标
  10. 后续行动： 战略下一步

Why This Works

为何该框架有效

  • Outcome-driven: Forces clarity on business AND customer value
  • Hypothesis-centric: Treats solution as a bet to validate, not a commitment
  • Risk-explicit: Makes assumptions and risks visible upfront
  • Executive-friendly: Comprehensive but structured for C-level review
  • AI-appropriate: Especially useful for AI features with high uncertainty
  • 成果导向： 明确要求同时聚焦业务和客户价值
  • 假设为中心： 将解决方案视为需要验证的赌注，而非既定承诺
  • 风险透明： 提前暴露假设和风险
  • 适配高管视角： 内容全面且结构清晰，适合高管层审阅
  • 适配AI场景： 尤其适用于具有高度不确定性的AI功能

Anti-Patterns (What This Is NOT)

反模式（该框架不是什么）

  • Not a PRD: This is strategic framing, not detailed requirements
  • Not a business case (yet): It informs the business case but needs validation first
  • Not a feature list: Focus on outcomes, not capabilities
  • 不是PRD： 这是战略层面的界定，而非详细需求说明
  • 暂不是商业案例： 它为商业案例提供信息，但需先经过验证
  • 不是功能列表： 聚焦成果，而非功能

When to Use This

适用场景

  • Proposing a new AI-powered product or feature
  • Pitching to execs or securing budget/sponsorship
  • Evaluating whether an AI solution is worth pursuing
  • Aligning cross-functional stakeholders (product, engineering, data science, business)
  • After completing initial discovery (you need context to fill this out)
  • 提出新的AI驱动产品或功能
  • 向高管汇报或争取预算/赞助
  • 评估AI解决方案是否值得推进
  • 协调跨职能利益相关者（产品、工程、数据科学、业务）
  • 完成初步探索工作后（你需要相关背景信息来填写此画布）

When NOT to Use This

不适用场景

  • For trivial features (don't over-engineer small tweaks)
  • Before any discovery work (you need user research and problem validation first)
  • As a replacement for experimentation (canvas informs experiments, not vice versa)
  • 针对琐碎功能（无需为小调整过度设计）
  • 未开展任何探索工作前（你需要先完成用户研究和问题验证）
  • 替代实验（画布为实验提供指导，而非反之）

Application

应用方法

Use 
template.md
for the full fill-in structure.
使用
template.md
获取完整的填写模板。

Step 1: Gather Context

步骤1：收集背景信息

Before filling out the canvas, ensure you have:
  • Problem understanding: User research, pain points (reference 
    skills/problem-statement/SKILL.md
    )
  • Persona clarity: Who experiences the problem? (reference 
    skills/proto-persona/SKILL.md
    )
  • Market context: Competitive landscape, category positioning
  • Business constraints: Budget, timelines, strategic priorities
If missing context: Run discovery work first. This canvas synthesizes insights—it doesn't create them.
在填写画布前，确保你已具备：
  • 问题理解： 用户研究痛点（参考
    skills/problem-statement/SKILL.md
    ）
  • 用户角色清晰： 谁在面临该问题？（参考
    skills/proto-persona/SKILL.md
    ）
  • 市场背景： 竞争格局、品类定位
  • 业务约束： 预算、时间线、战略优先级
若缺少背景信息： 先开展探索工作。该画布用于整合洞察——而非生成洞察。

Step 2: Define Outcomes

步骤2：界定成果

Business Outcome

业务成果

What's in it for the business? Use this format:
  • [Direction] [Metric] [Outcome] [Context] [Acceptance Criteria]
markdown
undefined
对业务有何价值？使用以下格式：
  • [方向] [指标] [成果] [背景] [验收标准]
markdown
undefined

Business Outcome

Business Outcome

  • [e.g., "Reduce by 25% the churn of existing customers using our existing product"]

**Example:**
- "Increase by 15% the monthly recurring revenue from enterprise customers within 12 months"

**Quality checks:**
- **Measurable:** Can you track this metric?
- **Time-bound:** Within what timeframe?
- **Ambitious but realistic:** Not "10x revenue in 1 month"

---
  • [e.g., "Reduce by 25% the churn of existing customers using our existing product"]

**示例：**
- "在12个月内将企业客户的月度经常性收入提升15%"

**质量检查：**
- **可衡量：** 你能否追踪该指标？
- **有时限：** 在什么时间范围内？
- **有野心但现实：** 不是“1个月内收入增长10倍”

---

Product Outcome

产品成果

What's in it for the customer? Use this format:
  • [Direction] [Metric] [Outcome] [Context from persona's POV] [Acceptance Criteria]
markdown
undefined
对客户有何价值？使用以下格式：
  • [方向] [指标] [成果] [从用户角色视角出发的背景] [验收标准]
markdown
undefined

Product Outcome

Product Outcome

  • [e.g., "Increase the speed of finding patients when I know the inclusion and exclusion criteria"]

**Example:**
- "Reduce by 60% the time spent manually processing invoices for small business owners"

**Quality checks:**
- **Customer-centric:** Written from user perspective ("I," not "we")
- **Outcome, not feature:** "Reduce time spent" not "Use AI automation"

---
  • [e.g., "Increase the speed of finding patients when I know the inclusion and exclusion criteria"]

**示例：**
- "为小企业主减少60%的发票手动处理时间"

**质量检查：**
- **以客户为中心：** 从用户视角撰写（使用“我”而非“我们”）
- **聚焦成果，而非功能：** “减少耗时”而非“使用AI自动化”

---

Step 3: Frame the Problem

步骤3：界定问题

Use the problem framing narrative from 
skills/problem-statement/SKILL.md
:
markdown
undefined
使用
skills/problem-statement/SKILL.md
中的问题界定叙事格式：
markdown
undefined

The Problem Statement

The Problem Statement

Problem Statement Narrative

Problem Statement Narrative

  • [Persona description: 2-3 sentences telling the persona's story from their POV]
  • [Example: "Sarah is a freelance designer managing 10 clients. She spends 8 hours/month manually tracking invoices and chasing late payments. By the time she follows up, some clients have already moved to other designers, costing her revenue and damaging relationships."]

**Quality checks:**
- **Empathetic:** Does this sound like the user's voice?
- **Specific:** Not "users want better tools" but "Sarah spends 8 hours/month..."
- **Validated:** Based on real user research, not assumptions

---
  • [Persona description: 2-3 sentences telling the persona's story from their POV]
  • [Example: "Sarah is a freelance designer managing 10 clients. She spends 8 hours/month manually tracking invoices and chasing late payments. By the time she follows up, some clients have already moved to other designers, costing her revenue and damaging relationships."]

**质量检查：**
- **共情：** 这听起来像用户的真实心声吗？
- **具体：** 不是“用户需要更好的工具”而是“Sarah每月花费8小时……”
- **已验证：** 基于真实用户研究，而非假设

---

Step 4: Define the Solution Hypothesis

步骤4：界定解决方案假设

Hypothesis Statement

假设陈述

Use the epic hypothesis format from 
skills/epic-hypothesis/SKILL.md
:
markdown
undefined
使用
skills/epic-hypothesis/SKILL.md
中的史诗假设格式：
markdown
undefined

Solution Hypothesis

Solution Hypothesis

Hypothesis Statement

Hypothesis Statement

If we [action or solution on behalf of target persona] for [target persona] Then we will [attain or achieve desirable outcome]

**Example:**
- "If we provide AI-powered invoice reminders that auto-send at optimal times for freelance designers, then we will reduce time spent on payment follow-ups by 70%"

---
If we [action or solution on behalf of target persona] for [target persona] Then we will [attain or achieve desirable outcome]

**示例：**
- "如果我们为自由设计师提供可在最佳时间自动发送的AI驱动发票提醒，那么我们将把付款跟进耗时减少70%"

---

Tiny Acts of Discovery

小型探索实验

Define lightweight experiments to validate the hypothesis:
markdown
undefined
定义轻量级实验以验证假设：
markdown
undefined

Tiny Acts of Discovery

Tiny Acts of Discovery

We will test our assumption by:
  • [Experiment 1: Prototype AI reminder system and test with 5 freelancers]
  • [Experiment 2: A/B test manual vs. AI-timed reminders for 20 users]
  • [Experiment 3: Survey users on perceived value after 2 weeks]

**Quality checks:**
- **Fast:** Days/weeks, not months
- **Cheap:** Prototypes, concierge tests, not full builds
- **Falsifiable:** Could prove you wrong

---
We will test our assumption by:
  • [Experiment 1: Prototype AI reminder system and test with 5 freelancers]
  • [Experiment 2: A/B test manual vs. AI-timed reminders for 20 users]
  • [Experiment 3: Survey users on perceived value after 2 weeks]

**质量检查：**
- **快速：** 数天/数周，而非数月
- **低成本：** 原型、礼宾式测试，而非完整开发
- **可证伪：** 有可能证明假设错误

---

Proof-of-Life

可行性验证

Define validation measures:
markdown
undefined
定义验证指标：
markdown
undefined

Proof-of-Life

Proof-of-Life

We know our hypothesis is valid if within [timeframe] we observe:
  • [Quantitative outcome: e.g., "80% of users send reminders via the AI system"]
  • [Qualitative outcome: e.g., "8 out of 10 users report saving 5+ hours/month"]

---
We know our hypothesis is valid if within [timeframe] we observe:
  • [Quantitative outcome: e.g., "80% of users send reminders via the AI system"]
  • [Qualitative outcome: e.g., "8 out of 10 users report saving 5+ hours/month"]

---

Step 5: Define Positioning

步骤5：界定定位

Use the positioning statement format from 
skills/positioning-statement/SKILL.md
:
markdown
undefined
使用
skills/positioning-statement/SKILL.md
中的定位陈述格式：
markdown
undefined

Positioning Statement

Positioning Statement

Value Proposition

Value Proposition

For [target customer/user persona] that need [statement of underserved need] [product name] is a [product category] that [statement of benefit, focusing on outcomes]
For [target customer/user persona] that need [statement of underserved need] [product name] is a [product category] that [statement of benefit, focusing on outcomes]

Differentiation Statement

Differentiation Statement

Unlike [primary competitor or competitive arena] [product name] provides [unique differentiation, focusing on outcomes]

---
Unlike [primary competitor or competitive arena] [product name] provides [unique differentiation, focusing on outcomes]

---

Step 6: Document Assumptions & Unknowns

步骤6：记录假设与未知因素

markdown
undefined
markdown
undefined

Assumptions & Unknowns

Assumptions & Unknowns

  • [Assumption 1] - [Description, e.g., "We assume users will trust AI-generated reminders"]
  • [Assumption 2] - [Description, e.g., "We assume payment timing optimization increases response rates"]
  • [Unknown 1] - [Description, e.g., "We don't know if users prefer email or SMS reminders"]

**Quality checks:**
- **Explicit:** Make hidden assumptions visible
- **Testable:** Each assumption can be validated via experiments

---
  • [Assumption 1] - [Description, e.g., "We assume users will trust AI-generated reminders"]
  • [Assumption 2] - [Description, e.g., "We assume payment timing optimization increases response rates"]
  • [Unknown 1] - [Description, e.g., "We don't know if users prefer email or SMS reminders"]

**质量检查：**
- **明确：** 让隐藏的假设显性化
- **可测试：** 每个假设都可通过实验验证

---

Step 7: Identify PESTEL Risks

步骤7：识别PESTEL风险

Risks to Investigate (High Priority)

需重点调查的风险（高优先级）

markdown
undefined
markdown
undefined

Issues/Risks to Investigate

Issues/Risks to Investigate

  • Political: [e.g., "Regulatory changes to AI-generated communications"]
  • Economic: [e.g., "Economic downturn reduces willingness to pay for premium features"]
  • Social: [e.g., "Users may perceive AI reminders as impersonal or pushy"]
  • Technological: [e.g., "AI model accuracy may degrade over time without retraining"]
  • Environmental: [e.g., "Energy costs of AI processing"]
  • Legal: [e.g., "GDPR compliance for storing customer email patterns"]

---
  • Political: [e.g., "Regulatory changes to AI-generated communications"]
  • Economic: [e.g., "Economic downturn reduces willingness to pay for premium features"]
  • Social: [e.g., "Users may perceive AI reminders as impersonal or pushy"]
  • Technological: [e.g., "AI model accuracy may degrade over time without retraining"]
  • Environmental: [e.g., "Energy costs of AI processing"]
  • Legal: [e.g., "GDPR compliance for storing customer email patterns"]

---

Risks to Monitor (Lower Priority)

需监控的风险（低优先级）

markdown
undefined
markdown
undefined

Issues/Risks to Monitor

Issues/Risks to Monitor

  • Political: [e.g., "Potential AI regulation in EU markets"]
  • Economic: [e.g., "Exchange rate fluctuations affecting international customers"]
  • Social: [e.g., "Changing norms around automated communication"]
  • Technological: [e.g., "Emerging AI competitors with better models"]
  • Environmental: [e.g., "Carbon footprint concerns from stakeholders"]
  • Legal: [e.g., "Future data privacy laws"]

---
  • Political: [e.g., "Potential AI regulation in EU markets"]
  • Economic: [e.g., "Exchange rate fluctuations affecting international customers"]
  • Social: [e.g., "Changing norms around automated communication"]
  • Technological: [e.g., "Emerging AI competitors with better models"]
  • Environmental: [e.g., "Carbon footprint concerns from stakeholders"]
  • Legal: [e.g., "Future data privacy laws"]

---

Step 8: Justify the Value

步骤8：价值合理性论证

markdown
undefined
markdown
undefined

Value Justification

Value Justification

Is this Valuable?

Is this Valuable?

  • [Absolutely yes / Yes with caveats / No with suggested alternatives / Absolutely NO!]
  • [Absolutely yes / Yes with caveats / No with suggested alternatives / Absolutely NO!]

Solution Justification

Solution Justification

<!-- Write these to convince C-level executives -->
We think this is a valuable idea. Here's why:
  1. [Justification 1] - [Description, e.g., "Addresses the #1 pain point for our target segment"]
  2. [Justification 2] - [Description, e.g., "Differentiates us from competitors who only offer manual reminders"]
  3. [Justification 3] - [Description, e.g., "Low technical risk—leverages existing AI infrastructure"]

---
<!-- Write these to convince C-level executives -->
We think this is a valuable idea. Here's why:
  1. [Justification 1] - [Description, e.g., "Addresses the #1 pain point for our target segment"]
  2. [Justification 2] - [Description, e.g., "Differentiates us from competitors who only offer manual reminders"]
  3. [Justification 3] - [Description, e.g., "Low technical risk—leverages existing AI infrastructure"]

---

Step 9: Define Success Metrics

步骤9：定义成功指标

Use SMART metrics (Specific, Measurable, Attainable, Relevant, Time-Bound):
markdown
undefined
使用SMART指标（具体、可衡量、可实现、相关、有时限）：
markdown
undefined

Success Metrics

Success Metrics

  1. [Metric 1] - [e.g., "80% of active users adopt AI reminders within 3 months"]
  2. [Metric 2] - [e.g., "Average time spent on payment follow-ups decreases by 50% within 6 months"]
  3. [Metric 3] - [e.g., "Net Promoter Score for invoicing feature increases from 6 to 8 within 6 months"]

---
  1. [Metric 1] - [e.g., "80% of active users adopt AI reminders within 3 months"]
  2. [Metric 2] - [e.g., "Average time spent on payment follow-ups decreases by 50% within 6 months"]
  3. [Metric 3] - [e.g., "Net Promoter Score for invoicing feature increases from 6 to 8 within 6 months"]

---

Step 10: Define Next Steps

步骤10：定义后续行动

markdown
undefined
markdown
undefined

What's Next

What's Next

  1. [Next step 1] - [e.g., "Run 2-week prototype test with 10 beta users"]
  2. [Next step 2] - [e.g., "Build lightweight AI model for reminder timing optimization"]
  3. [Next step 3] - [e.g., "Conduct legal review of GDPR implications"]
  4. [Next step 4] - [e.g., "Present findings to exec team for go/no-go decision"]
  5. [Next step 5] - [e.g., "If validated, add to Q2 roadmap"]

---
  1. [Next step 1] - [e.g., "Run 2-week prototype test with 10 beta users"]
  2. [Next step 2] - [e.g., "Build lightweight AI model for reminder timing optimization"]
  3. [Next step 3] - [e.g., "Conduct legal review of GDPR implications"]
  4. [Next step 4] - [e.g., "Present findings to exec team for go/no-go decision"]
  5. [Next step 5] - [e.g., "If validated, add to Q2 roadmap"]

---

Examples

示例

See 
examples/sample.md
for a full recommendation canvas example.
Mini example excerpt:
markdown
undefined
查看
examples/sample.md
获取完整的推荐画布示例。
迷你示例摘录：
markdown
undefined

Business Outcome

Business Outcome

  • Increase by 20% MRR from freelance users within 12 months
  • Increase by 20% MRR from freelance users within 12 months

Solution Hypothesis

Solution Hypothesis

If we provide AI-powered invoice reminders for freelance designers Then we will reduce time spent on follow-ups by 70%
undefined
If we provide AI-powered invoice reminders for freelance designers Then we will reduce time spent on follow-ups by 70%
undefined

Common Pitfalls

常见陷阱

Pitfall 1: Vague Outcomes

陷阱1：模糊的成果

Symptom: "Business outcome: increase revenue. Product outcome: improve UX."
Consequence: No measurability or accountability.
Fix: Use the outcome formula: [Direction] [Metric] [Outcome] [Context] [Acceptance Criteria]. Be specific.
症状： “业务成果：增加收入。产品成果：提升UX。”
后果： 缺乏可衡量性和问责性。
解决方法： 使用成果公式：[方向] [指标] [成果] [背景] [验收标准]。做到具体明确。

Pitfall 2: Solution-First Thinking

陷阱2：先入为主的解决方案思维

Symptom: Problem statement is "We need AI-powered X"
Consequence: You've jumped to solution without validating the problem.
Fix: Frame problem from user perspective. Let the solution hypothesis emerge from validated pain points.
症状： 问题陈述为“我们需要AI驱动的X”
后果： 你在未验证问题的情况下就跳到了解决方案。
解决方法： 从用户视角界定问题。让解决方案假设从已验证的痛点中自然衍生。

Pitfall 3: Skipping Tiny Acts of Discovery

陷阱3：跳过小型探索实验

Symptom: Hypothesis → straight to roadmap, no experiments
Consequence: High risk of building the wrong thing.
Fix: Define 2-3 lightweight experiments. Test before committing engineering resources.
症状： 提出假设后直接进入路线图，未开展实验
后果： 构建错误产品的高风险。
解决方法： 定义2-3个轻量级实验。在投入工程资源前先进行测试。

Pitfall 4: Generic PESTEL Risks

陷阱4：通用的PESTEL风险

Symptom: "Political: regulations might change"
Consequence: Risk analysis is theater, not actionable.
Fix: Be specific: "GDPR compliance for storing client email timing data requires legal review."
症状： “政治：法规可能变化”
后果： 风险分析流于形式，不具备可操作性。
解决方法： 做到具体：“存储客户邮件时间数据的GDPR合规性需要法律审查。”

Pitfall 5: Weak Value Justification

陷阱5：薄弱的价值合理性论证

Symptom: "This is valuable because customers will like it"
Consequence: Not convincing to execs.
Fix: Use data: "Addresses #1 pain point per user research. 20% churn reduction = $500k ARR. Low tech risk."
症状： “这很有价值，因为客户会喜欢它”
后果： 无法说服高管。
解决方法: 使用数据：“根据用户研究，这解决了目标群体的头号痛点。减少20%的客户流失相当于50万美元的年度经常性收入。技术风险低。”

References

参考资料

Related Skills

相关技能

  • skills/problem-statement/SKILL.md
    — Informs the problem narrative
  • skills/epic-hypothesis/SKILL.md
    — Informs the solution hypothesis structure
  • skills/positioning-statement/SKILL.md
    — Informs positioning section
  • skills/proto-persona/SKILL.md
    — Defines target persona
  • skills/jobs-to-be-done/SKILL.md
    — Informs customer outcomes
  • skills/problem-statement/SKILL.md
    — 为问题叙事提供信息
  • skills/epic-hypothesis/SKILL.md
    — 为解决方案假设结构提供信息
  • skills/positioning-statement/SKILL.md
    — 为定位部分提供信息
  • skills/proto-persona/SKILL.md
    — 定义目标用户角色
  • skills/jobs-to-be-done/SKILL.md
    — 为客户成果提供信息

External Frameworks

外部框架

  • Osterwalder's Value Proposition Canvas — Influences problem/solution framing
  • PESTEL Analysis — Risk assessment framework
  • SMART Goals — Success metrics structure
  • Osterwalder的价值主张画布 — 影响问题/解决方案界定
  • PESTEL分析 — 风险评估框架
  • SMART目标 — 成功指标结构

Dean's Work

Dean的作品

  • AI Recommendation Canvas Template (created for Productside "AI Innovation for Product Managers" class)
  • AI推荐画布模板（为Productside“产品经理的AI创新”课程创建）

Provenance

来源

  • Adapted from 
    prompts/recommendation-canvas-template.md
    in the 
    https://github.com/deanpeters/product-manager-prompts
    repo.
Skill type: Component Suggested filename: 
recommendation-canvas.md
Suggested placement: 
/skills/components/
Dependencies: References 
skills/problem-statement/SKILL.md
, 
skills/epic-hypothesis/SKILL.md
, 
skills/positioning-statement/SKILL.md
, 
skills/proto-persona/SKILL.md
, 
skills/jobs-to-be-done/SKILL.md
  • 改编自
    https://github.com/deanpeters/product-manager-prompts
    仓库中的
    prompts/recommendation-canvas-template.md
    。
技能类型： 组件 建议文件名： 
recommendation-canvas.md
建议存放位置： 
/skills/components/
依赖项： 参考
skills/problem-statement/SKILL.md
、
skills/epic-hypothesis/SKILL.md
、
skills/positioning-statement/SKILL.md
、
skills/proto-persona/SKILL.md
、
skills/jobs-to-be-done/SKILL.md