Preamble (run first)
# 检查是否有用户旅程地图作为输入
if [ -f "docs/01-需求调研/用户旅程地图.md" ]; then
echo "✅ 检测到用户旅程地图，将基于此进行漏斗分析"
fi
Write-time boundaries
When only discussing, do not write files; startup checks do not create directories. When the user asks to save an artifact, first confirm the current product, the exact output path, and the overwrite/append choice for existing files; after confirmationonly immediately before the actual Writecreate the parent directory of that output file. For full installation, see skills/_shared/artifact-policy.md, and installing this skill separately also follows this paragraph.
Cross-Agent interaction rules
When the process requires interaction with the user:
- If the current environment supports AskUserQuestion, use AskUserQuestion (best experience).
- If the current environment does not support AskUserQuestion, you must ask the same question via a normal chat message.
- Ask only one question at a time.
- After asking, you must stop the current turn and wait for the user's answer (STOP and WAIT).
- Do not generate documents or write to docs before the user answers.
- Existing docs files cannot replace the user's answer in this turn.
Execution flow
Step 1: Define the conversion funnel
How to ask: use AskUserQuestion to ask one by one
Question 1: Business scenario
Use AskUserQuestion to ask:
Which business scenario's conversion funnel do you want to analyze?
A) Registration conversion funnel (visitor→registered user) B) Purchase conversion funnel (browse→place order→pay) C) Activation conversion funnel (download→register→first use) D) Retention conversion funnel (new→next-day retention→7-day retention) E) Other (please enter manually)
Question 2: Funnel step definition
Based on the scenario in Question 1, guide the user to define the specific steps:
Example (purchase conversion funnel)：
Please confirm the key steps of the purchase conversion funnel
Guide the user to input:
- Step 1: [Browse product detail page]
- Step 2: [Add to cart]
- Step 3: [Enter checkout page]
- Step 4: [Submit order]
- Step 5: [Complete payment]
Question 3: Current data
For each step, collect the current data:
Use AskUserQuestion to ask:
What is the number of users for Step [1]: [Browse product detail page]?
Guide the user to input the number of users for each step:
| Step | Number of users | Percentage |
|---|---|---|
| Step 1 | [X] ten thousand | 100% |
| Step 2 | [X] ten thousand | [X]% |
| Step 3 | [X] ten thousand | [X]% |
| Step 4 | [X] thousand | [X]% |
| Step 5 | [X] thousand | [X]% |
Step 2: Calculate conversion rate and churn rate
Calculation formula：
步骤转化率 = (下一步用户数 / 当前步骤用户数) × 100%
步骤流失率 = 100% - 步骤转化率
整体转化率 = (最后一步用户数 / 第一步用户数) × 100%
Example：
| Step | Number of users | Step conversion rate | Step churn rate |
|---|---|---|---|
| Browse detail page | 10,000 | - | - |
| Add to cart | 3,000 | 30% | 70% |
| Enter checkout page | 2,100 | 70% | 30% |
| Submit order | 1,470 | 70% | 30% |
| Complete payment | 882 | 60% | 40% |
Overall conversion rate：882 / 10,000 = 8.82%
Step 3: Churn Cause Diagnosis
For the step with the highest churn rate, analyze the causes:
Step 3.1: Identify Key Churn Nodes
Definition of Key Churn Nodes: Steps with churn rate > 30%
Example：
- Step 1→2: Churn rate 70% (key churn node)
- Step 4→5: Churn rate 40% (key churn node)
Step 3.2: Diagnose Churn Causes
For key churn nodes, use AskUserQuestion to ask:
Step 1→2 (Browse detail page→Add to cart) has a churn rate as high as 70%. What are the possible reasons?
A) Product price too high B) Product information unclear C) User purchase intent not strong D) Churn due to competitor comparison E) Page loads slowly F) Other (please enter manually)
For each churn cause, guide the user to analyze in depth:
Please provide data support or user feedback (if any)
Step 3.3: User Feedback Collection
Ask:
Is there user research or customer service feedback data?
A) Have user research data B) Have customer service feedback records C) Have user reviews/complaints D) No data yet, need to collect
If D is selected, provide data collection suggestions:
Suggested Collection Methods：
- User questionnaire survey (churned user questionnaire)
- Customer service feedback analysis
- User interviews (in-depth interviews with 5-10 people)
- Data tracking (detailed behavioral data)
Step 4: Generate Optimization Suggestions
Based on the churn causes, generate targeted optimization suggestions:
Scenario A: Churn caused by product price being too high
Optimization Suggestions：
- Provide price discounts (first-order discount, coupons)
- Show value for money (compare with competitors, highlight advantages)
- Installment payment options
- Price anchoring strategy (show high-priced product comparisons)
Scenario B: Churn caused by unclear product information
Optimization Suggestions：
- Optimize product detail page design
- Add video displays
- Move user reviews to the front
- Highlight key information
Scenario C: Churn caused by slow page loading
Optimization Suggestions：
- Optimize image loading (compression, lazy loading)
- CDN acceleration
- Server performance optimization
- Reduce unnecessary requests
Scenario D: Churn caused by complex checkout process
Optimization Suggestions：
- Simplify checkout steps
- One-click purchase feature
- Default address/payment method
- Progress bar guidance
Step 5: A/B Testing Suggestions
For key optimization items, design A/B tests:
Test Design Template：
测试名称：[优化项目名称]
测试目标：提升步骤[X]到步骤[X+1]的转化率
测试周期：[X]周
样本量：[X]万用户
A组（对照组）：[当前方案]
B组（实验组）：[优化方案]
评估指标：
- 主指标：步骤转化率
- 辅助指标：整体转化率、GMV
预期效果：转化率提升[X]%
Step 6: Generate Funnel Optimization Plan
Use the Write tool to generate a document to docs/05-产品战略/漏斗优化方案.md：
---
product: [产品名称]
created_at: [当前时间]
author: [用户]
skill: pm-funnel
status: draft
---
# 漏斗分析与优化方案
## 一、漏斗概览
### 业务场景
**分析场景**：购买转化漏斗
**分析周期**：[起始日期] - [结束日期]
**数据来源**：[数据平台名称]
---
### 漏斗定义
**转化路径**：
1. 浏览商品详情页
2. 加入购物车
3. 进入结算页
4. 提交订单
5. 完成支付
---
## 二、漏斗数据分析
### 整体数据
| 步骤 | 用户数 | 步骤转化率 | 步骤流失率 | 累计转化率 |
|------|-------|-----------|-----------|-----------|
| 浏览详情页 | 10,000 | - | - | 100% |
| 加入购物车 | 3,000 | 30% | 70% | 30% |
| 进入结算页 | 2,100 | 70% | 30% | 21% |
| 提交订单 | 1,470 | 70% | 30% | 14.7% |
| 完成支付 | 882 | 60% | 40% | 8.82% |
**关键指标**：
- 整体转化率：8.82%
- 平均步骤转化率：57.5%
- 最大流失节点：浏览详情页→加入购物车（流失率70%）
---
### 流失用户画像
**流失用户特征**：
- 新用户占比：[X]%
- 来源渠道：[渠道分布]
- 设备分布：[iOS/Android占比]
- 访问时段：[时段分布]
---
## 三、关键流失节点分析
### 节点1: 浏览详情页→加入购物车
**流失率**：70%
**流失用户数**：7,000人
**流失原因诊断**：
| 原因 | 占比 | 数据支撑 |
|------|------|---------|
| 价格高于预期 | 35% | 用户调研：65%用户认为价格偏高 |
| 商品信息不全 | 25% | 客服反馈：商品描述不清 |
| 页面加载慢 | 20% | 平均加载时间：5.2秒 |
| 竞品对比流失 | 15% | 跳出用户：40%访问竞品 |
| 其他原因 | 5% | - |
**详细分析**：
1. **价格因素**（35%流失）
- 商品平均价格：[X]元
- 用户期望价格：[X]元
- 竞品平均价格：[X]元
- 价格敏感度：高
2. **信息因素**（25%流失）
- 商品图片数量：平均3张
- 商品描述字数：平均150字
- 视频展示：缺失
- 用户评价：平均15条
3. **性能因素**（20%流失）
- 平均加载时间：5.2秒
- 行业标准：3秒以内
- 跳出率：[X]%
---
### 节点2: 提交订单→完成支付
**流失率**：40%
**流失用户数**：588人
**流失原因诊断**：
| 原因 | 占比 | 数据支撑 |
|------|------|---------|
| 支付方式有限 | 30% | 仅支持微信支付 |
| 结算流程复杂 | 25% | 平均结算时间：3分钟 |
| 运费过高 | 20% | 平均运费：15元 |
| 支付失败 | 15% | 支付成功率：85% |
| 其他原因 | 10% | - |
---
## 四、优化建议
### 优先级排序
| 优化项 | 预期提升 | 实施难度 | 投入成本 | 优先级 |
|-------|---------|---------|---------|--------|
| 价格策略优化 | +5% | 中 | 低 | P0 |
| 详情页优化 | +3% | 低 | 低 | P1 |
| 页面性能优化 | +2% | 中 | 中 | P1 |
| 支付流程优化 | +1.5% | 低 | 低 | P2 |
| 支付方式扩展 | +1% | 高 | 高 | P2 |
---
### 优化方案详情
#### 方案1: 价格策略优化（P0）
**目标**：降低价格敏感用户流失，提升转化率5%
**具体措施**：
1. 首单优惠：新用户首单立减[X]元
2. 满减活动：满[X]元减[X]元
3. 优惠券：定向发放优惠券
4. 价格锚定：展示原价vs现价
**实施计划**：
- Week 1：设计优惠券方案
- Week 2：开发优惠功能
- Week 3：上线测试
**预期效果**：
- 浏览→加购转化率：30% → 35%（+5%）
- 整体转化率：8.82% → 9.3%（+0.48%）
---
#### 方案2: 详情页优化（P1）
**目标**：提升信息完整性，降低信息缺失流失
**具体措施**：
1. 增加商品视频展示
2. 优化商品图片（增至6-8张）
3. 用户评价前置
4. 关键参数表格化展示
**实施计划**：
- Week 1：设计详情页模板
- Week 2：开发新模板
- Week 3：商家培训
**预期效果**：
- 浏览→加购转化率：30% → 33%（+3%）
---
#### 方案3: 页面性能优化（P1）
**目标**：提升加载速度，降低性能流失
**具体措施**：
1. 图片压缩（WebP格式）
2. CDN加速
3. 懒加载优化
4. 服务端渲染（SSR）
**实施计划**：
- Week 1：性能分析
- Week 2-3：优化实施
- Week 4：效果验证
**目标指标**：
- 平均加载时间：5.2秒 → 2.5秒
- 跳出率：[X]% → [X]%
---
#### 方案4: 支付流程优化（P2）
**目标**：简化支付流程，提升支付转化率
**具体措施**：
1. 一键支付功能
2. 默认地址记忆
3. 支付进度提示
4. 支付失败自动重试
**预期效果**：
- 提交→支付转化率：60% → 70%（+10%）
---
## 五、A/B测试计划
### 测试1: 价格优惠策略
**测试名称**：首单优惠效果测试
**测试周期**：2周
**样本量**：10,000新用户
**测试设计**：
- A组（对照组）：无优惠
- B组（实验组）：首单立减10元
**评估指标**：
- 主指标：浏览→加购转化率
- 辅助指标：整体转化率、GMV、客单价
**预期效果**：
- 转化率提升：+5%
- GMV提升：+3%
---
### 测试2: 详情页优化
**测试名称**：视频展示效果测试
**测试周期**：2周
**样本量**：20,000用户
**测试设计**：
- A组（对照组）：现有详情页
- B组（实验组）：增加视频展示
**评估指标**：
- 主指标：浏览→加购转化率
- 辅助指标：页面停留时长、跳出率
---
## 六、监控与迭代
### 监控指标
**实时监控**（每小时）：
- 各步骤转化率
- 整体转化率
- GMV
**周报监控**：
- 转化率趋势
- 流失原因变化
- 优化效果对比
---
### 迭代机制
**迭代周期**：每2周复盘一次
**迭代流程**：
1. 数据分析 → 识别问题
2. 假设提出 → 设计优化方案
3. A/B测试 → 验证效果
4. 全量上线 → 监控效果
5. 持续优化
---
## 七、下一步建议
建议执行：
1. /pm-abtest - 设计详细的A/B测试方案
2. /pm-growth - 制定增长执行方案
3. /pm-report - 建立数据监控体系
---
**项目状态**: 漏斗分析完成
**生成时间**: [当前时间]
**生成工具**: super-pm
Notes
- Data Accuracy: Ensure data tracking is accurate and avoid data bias
- Multi-dimensional Analysis: Combine multi-dimensional analysis such as user segmentation, channels, and devices
- Accurate Attribution: Correctly attribute churn causes and avoid misjudgment
- Continuous Optimization: Funnel optimization is a continuous process and requires regular review
Output Quality Comparison
✅ Good Example：
- 有数据引用：「根据 Q4 数据，留存率从 35% 降至 28%」
- 有验证来源：「数据来源：Google Analytics, 2025-12-01」
- 有明确建议：「建议将新手引导步骤从 5 步减少至 3 步」
❌ Bad Example：
- 模糊结论：「数据表明留存率有所下降」
- 无来源：「根据经验，这个功能很重要」
- 没有行动建议：「留存是个问题」
Common Pitfalls / Red Flags — STOP
Stop immediately and backtrack if any of the following occur:
| Pitfall | Correct Approach |
|---|---|
| Drawing conclusions using "should", "probably", "seems" | Must be based on actual data and verification |
| Claiming completion without running checks | Verify first, then state |
| Skipping critical steps due to time pressure | No exceptions; the tighter the time, the stricter you must be |
| The thought "this should be fine" | Re-verify every time |
Output Quality Check / Verification Checklist
- Prerequisite dependencies satisfied (input documents/data collected)
- All core steps executed
- Output document generated to
docs/directory - Every judgment supported by data/evidence
- 2-3 follow-up skills recommended
⚠️ If any item fails → complete it before marking as done.