---

name: analytics-product description: "产品分析——PostHog、Mixpanel、事件、漏斗、群组、留存、北极星指标、OKR 和产品仪表板。" risk: none source: community date_added: '2026-03-06' author: renat tags:

• analytics
• product
• metrics
• posthog
• mixpanel tools:
• claude-code
• antigravity
• cursor
• gemini-cli
• codex-cli

---

ANALYTICS-PRODUCT — 用数据做决策

概述

产品分析——PostHog、Mixpanel、事件、漏斗、群组、留存、北极星指标、OKR 和产品仪表板。激活场景：配置事件追踪、创建转化漏斗、群组分析、留存分析、DAU/MAU、功能标志、A/B 测试、北极星指标、OKR、产品仪表板。

何时使用此技能

• 当你需要此领域的专业帮助时

不使用此技能的情况

• 任务与产品分析无关
• 更简单、更具体的工具可以处理请求
• 用户需要无领域专业知识的通用帮助

工作原理

[对象]_[过去式动词]

正确：   user_signed_up, conversation_started, upgrade_completed
错误：   signup, click, conversion

产品分析——用数据做决策

"In God we trust. All others must bring data." — W. Edwards Deming

---

Auri 核心事件

AURI_EVENTS = {
    # 获客
    "user_signed_up":        {"props": ["source", "medium", "campaign"]},
    "onboarding_started":    {"props": ["step_count"]},
    "onboarding_completed":  {"props": ["time_to_complete", "steps_skipped"]},

    # 激活
    "first_conversation":    {"props": ["intent", "response_time"]},
    "aha_moment_reached":    {"props": ["trigger", "session_number"]},
    "feature_discovered":    {"props": ["feature_name", "discovery_method"]},

    # 留存
    "conversation_started":  {"props": ["intent", "user_tier", "device"]},
    "conversation_completed":{"props": ["messages_count", "duration", "rating"]},
    "session_started":       {"props": ["days_since_last", "platform"]},

    # 收入
    "upgrade_viewed":        {"props": ["trigger", "current_tier"]},
    "upgrade_started":       {"props": ["target_tier", "trigger"]},
    "upgrade_completed":     {"props": ["tier", "plan", "revenue"]},
    "subscription_canceled": {"props": ["reason", "tier", "tenure_days"]},
    "payment_failed":        {"props": ["attempt_count", "error_code"]},
}

PostHog 实现（Python）

from posthog import Posthog
import os

posthog = Posthog(
    project_api_key=os.environ["POSTHOG_API_KEY"],
    host=os.environ.get("POSTHOG_HOST", "https://app.posthog.com")
)

def track(user_id: str, event: str, properties: dict = None):
    posthog.capture(
        distinct_id=user_id,
        event=event,
        properties=properties or {}
    )

def identify(user_id: str, traits: dict):
    posthog.identify(
        distinct_id=user_id,
        properties=traits
    )

## 用法：

track("user_123", "conversation_started", {
    "intent": "business_advice",
    "device": "alexa",
    "user_tier": "pro"
})

---

Auri 激活漏斗

访问落地页                    (100%)
    | [目标: 40%]
点击"立即体验"                (40%)
    | [目标: 70%]
完成注册                      (28%)
    | [目标: 60%]
进行第一次对话                (17%)  <- 顿悟时刻
    | [目标: 50%]
次日回访                      (8.5%)
    | [目标: 40%]
每周使用 3+ 天                (3.4%)
    | [目标: 20%]
转化为 Pro                    (0.7%)

优化漏斗

对于每个超过基准的流失点：
1. 识别：用户具体在哪里离开？
2. 理解：为什么？（会话录制、调查）
3. 假设：什么改变可能改善？
4. 测试：具有统计显著性样本的 A/B 测试
5. 衡量：最少 2 周，p-value < 0.05
6. 学习：即使失败，也能更好地理解用户

---

群组分析（周留存）

def calculate_cohort_retention(events_df):
    """
    events_df: 包含列 [user_id, event_date, event_name] 的 DataFrame
    返回：留存矩阵 [群组周 x 周数]
    """
    import pandas as pd

    first_session = events_df[events_df.event_name == "session_started"] \
        .groupby("user_id")["event_date"].min() \
        .dt.to_period("W")

    sessions = events_df[events_df.event_name == "session_started"].copy()
    sessions["cohort"] = sessions["user_id"].map(first_session)
    sessions["weeks_since"] = (
        sessions["event_date"].dt.to_period("W") - sessions["cohort"]
    ).apply(lambda x: x.n)

    cohort_data = sessions.groupby(["cohort", "weeks_since"])["user_id"].nunique()
    cohort_sizes = cohort_data.unstack().iloc[:, 0]
    retention = cohort_data.unstack().divide(cohort_sizes, axis=0) * 100

    return retention

留存基准（语音助手）

| 周 | 差 | 一般 | 好 | 优秀 | |--------|---------|-----|-----|-----------| | W1 | <20% | 20-35% | 35-50% | >50% | | W4 | <5% | 5-15% | 15-25% | >25% |