sales-pendo
Compare original and translation side by side
🇺🇸
Original
English🇨🇳
Translation
ChinesePendo Platform Help
Pendo平台帮助
Step 1 — Gather context
步骤1 — 收集背景信息
If exists, read it first for accumulated platform knowledge.
references/learnings.md-
What do you need help with?
- A) Setting up Pendo (install script, tagging, metadata)
- B) Product Analytics (funnels, paths, retention, feature adoption)
- C) In-App Guides (tooltips, modals, walkthroughs, Resource Center)
- D) Session Replay
- E) Feedback & Surveys (NPS, CSAT, PMF, polls)
- F) Predict (churn risk, upsell)
- G) Orchestrate (multi-channel coordination)
- H) API / Aggregation queries / Data Sync
- I) Integrations (HubSpot, Salesforce, Segment)
- J) Pricing / plan comparison
- K) Something else — describe it
-
What type of app?
- A) Web app (SaaS)
- B) Mobile app (iOS/Android)
- C) Both
-
What Pendo plan are you on? (Free / Base / Core / Ultimate / Not sure)
Skip-ahead rule: if the user's prompt already contains enough context, skip to Step 2.
如果文件存在,请先阅读该文件获取已积累的平台知识。
references/learnings.md-
你需要哪方面的帮助?
- A) 设置Pendo(安装脚本、标记、元数据)
- B) 产品分析(漏斗分析、用户路径、留存率、功能采用)
- C) 应用内引导(提示框、模态框、分步引导、资源中心)
- D) 会话重放
- E) 反馈与调查(NPS、CSAT、PMF、投票)
- F) Predict(流失风险、向上销售)
- G) Orchestrate(多渠道协调)
- H) API / 聚合查询 / 数据同步
- I) 集成(HubSpot、Salesforce、Segment)
- J) 定价/方案对比
- K) 其他——请描述具体内容
-
你的应用类型是什么?
- A) Web应用(SaaS)
- B) 移动应用(iOS/Android)
- C) 两者都是
-
你使用的是Pendo哪个方案?(免费版/基础版/核心版/旗舰版/不确定)
跳过规则:如果用户的提问已包含足够背景信息,直接进入步骤2。
Step 2 — Route or answer directly
步骤2 — 转介或直接解答
| Problem domain | Route to |
|---|---|
| In-app messaging strategy (not Pendo-specific) | |
| General NPS/CSAT/VoC program design | |
| Customer success health scores, churn prevention | |
| Connecting Pendo to CRM via Zapier/Make | |
If the question is Pendo-specific, answer directly using the reference below.
| 问题领域 | 转至 |
|---|---|
| 应用内消息策略(非Pendo专属) | |
| 通用NPS/CSAT/客户声音(VoC)方案设计 | |
| 客户成功健康评分、流失预防 | |
| 通过Zapier/Make将Pendo连接至CRM | |
如果问题是Pendo专属的,请使用下方参考内容直接解答。
Step 3 — Pendo platform reference
步骤3 — Pendo平台参考资料
Read for the full platform reference — modules, pricing, integrations, data model, workflows.
references/platform-guide.mdIf the question involves the API, also read .
references/pendo-api-reference.mdAnswer the user's question using only the relevant section. Don't dump the full reference.
请阅读获取完整平台参考——模块、定价、集成、数据模型、工作流。
references/platform-guide.md如果问题涉及API,还需阅读。
references/pendo-api-reference.md仅使用相关章节内容解答用户问题,不要直接粘贴完整参考资料。
Step 4 — Actionable guidance
步骤4 — 可落地的指导建议
Focus on the user's specific situation. Common patterns:
- Setup: Install script placement, visitor/account ID strategy, metadata schema design, page/feature tagging plan
- Analytics: Funnel construction, path analysis, retention cohorts, feature adoption tracking
- Guides: Targeting rules, activation triggers, guide chaining, A/B testing guides
- Feedback: Survey timing, throttling, closed-loop with CS tools
- API: Aggregation query construction, ETL patterns, metadata bulk updates
If you discover a gotcha, workaround, or tip not covered in , append it there.
references/learnings.md聚焦用户的具体场景。常见场景模式:
- 设置:安装脚本放置位置、访客/账户ID策略、元数据架构设计、页面/功能标记计划
- 分析:漏斗构建、路径分析、留存群组、功能采用追踪
- 引导:目标受众规则、激活触发条件、引导串联、引导A/B测试
- 反馈:调查时机、限流设置、与客户成功工具形成闭环
- API:聚合查询构建、ETL模式、元数据批量更新
如果发现中未涵盖的陷阱、变通方案或技巧,请将其添加至该文件。
references/learnings.mdGotchas
注意事项
Best-effort from research — review these, especially items about plan-gated features and integration gotchas that may be outdated.
- Free plan caps at 500 MAU. Above that you need Base or higher — no public pricing, expect $15K+/year.
- HubSpot and Salesforce integrations are paid add-ons. Not included in any standard plan — budget separately.
- Retroactive analytics only works for tagged pages/features. Pendo captures raw events automatically, but you can only analyze them retroactively once you tag the pages/features. Tag early.
- Guide targeting uses CSS selectors. If your app redesigns a page, guide anchors break. Use stable selectors (data attributes) instead of auto-generated class names.
- Aggregation API uses MongoDB-style pipelines. The query syntax is powerful but unfamiliar to most teams — expect a learning curve.
- Session Replay requires Core plan or higher (or as a paid add-on on Base). Not available on Free.
- Metadata must be pre-defined in the UI before you can send custom visitor/account properties via the install script or API.
- No native email channel. Pendo only reaches users inside your app. For dormant users, pair with an email tool.
基于研究的最佳实践——请仔细查看这些内容,尤其是关于方案限制功能和集成陷阱的内容,这些信息可能已过时。
- 免费版上限为500月活跃用户(MAU)。超过该数量需使用基础版或更高版本——无公开定价,预计每年费用15000美元以上。
- HubSpot和Salesforce集成是付费附加功能。不包含在任何标准方案中——需单独预算。
- 回溯分析仅适用于已标记的页面/功能。Pendo会自动捕获原始事件,但只有在标记页面/功能后才能对其进行回溯分析。尽早标记。
- 引导定位使用CSS选择器。如果你的应用重新设计了页面,引导锚点会失效。请使用稳定的选择器(数据属性)而非自动生成的类名。
- 聚合API使用MongoDB风格的管道。查询语法功能强大,但大多数团队不熟悉——需要学习成本。
- 会话重放需要核心版或更高版本(或作为基础版的付费附加功能)。免费版不可用。
- 元数据必须先在UI中预定义,才能通过安装脚本或API发送自定义访客/账户属性。
- 无原生邮件渠道。Pendo仅能触达应用内的用户。对于休眠用户,请搭配邮件工具使用。
Related skills
相关技能
- — In-app messaging strategy across platforms (Pendo, Appcues, Braze, Iterable). Install:
/sales-in-app-messagingnpx skills add sales-skills/sales --skill sales-in-app-messaging - — NPS, CSAT, VoC strategy and platform comparison. Install:
/sales-customer-feedbacknpx skills add sales-skills/sales --skill sales-customer-feedback - — Customer success strategy: health scores, churn prevention, onboarding. Install:
/sales-customer-successnpx skills add sales-skills/sales --skill sales-customer-success - — Not sure which skill to use? The router matches any sales objective to the right skill. Install:
/sales-donpx skills add sales-skills/sales --skill sales-do
- — 跨平台应用内消息策略(Pendo、Appcues、Braze、Iterable)。安装方式:
/sales-in-app-messagingnpx skills add sales-skills/sales --skill sales-in-app-messaging - — NPS、CSAT、VoC策略及平台对比。安装方式:
/sales-customer-feedbacknpx skills add sales-skills/sales --skill sales-customer-feedback - — 客户成功策略:健康评分、流失预防、新用户引导。安装方式:
/sales-customer-successnpx skills add sales-skills/sales --skill sales-customer-success - — 不确定使用哪个技能?该路由会将任何销售目标匹配到合适的技能。安装方式:
/sales-donpx skills add sales-skills/sales --skill sales-do
Examples
示例
Example 1: Feature adoption is low after launch
示例1:功能发布后采用率低
User says: "We shipped a new dashboard feature two weeks ago but Pendo shows only 12% of users have tried it. How do I increase adoption?"
Skill does: Analyzes the feature's Pendo analytics (paths, funnels from entry points), designs a targeted in-app guide campaign: tooltip on the nav item for users who haven't visited the feature, a walkthrough guide for first-time visitors, and a banner announcement for the broader user base. Includes targeting rules, display frequency, and success metrics.
Result: Multi-touch guide campaign with targeting, plus analytics setup to measure before/after adoption rate.
用户提问:"我们两周前发布了一个新的仪表盘功能,但Pendo显示只有12%的用户尝试过它。如何提高采用率?"
技能处理:分析该功能的Pendo分析数据(路径、从入口点开始的漏斗),设计针对性的应用内引导活动:向未访问过该功能的用户展示导航项上的提示框,为首次访问者提供分步引导,向更广泛的用户群体展示横幅公告。包含目标受众规则、展示频率和成功指标。
结果:带有目标定位的多触点引导活动,以及用于衡量采用率前后变化的分析设置。
Example 2: Pendo aggregation query for monthly active features
示例2:用于月度活跃功能的Pendo聚合查询
User says: "I need to pull feature usage data from Pendo into our data warehouse for executive reporting."
Skill does: Walks through the Aggregation API: endpoint (), authentication header, query structure with source, date filtering with , and grouping by feature and account. Provides a working JSON query example and suggests scheduling via cron or an ETL tool.
Result: Working aggregation query with authentication setup and ETL integration pattern.
POST /api/v1/aggregationfeatureEventstimeSeries用户提问:"我需要将Pendo的功能使用数据提取到我们的数据仓库,用于高管报告。"
技能处理:逐步讲解聚合API:端点()、认证头、使用数据源的查询结构、通过进行日期筛选,以及按功能和账户分组。提供可用的JSON查询示例,并建议通过cron或ETL工具进行调度。
结果:可用的聚合查询,包含认证设置和ETL集成模式。
POST /api/v1/aggregationfeatureEventstimeSeriesExample 3: Setting up Pendo for the first time
示例3:首次设置Pendo
User says: "I just got Pendo and need to install it on our React SaaS app. Where do I start?"
Skill does: Covers the install script snippet placement, visitor ID and account ID strategy (use your internal user ID, not email), metadata schema design (role, plan, signup date), initial page and feature tagging plan, and first guide creation. Flags common mistakes: using email as visitor ID (breaks on email changes), not pre-defining metadata fields in the UI, and over-tagging features on day one.
Result: Complete setup plan with install script, metadata schema, and tagging strategy.
用户提问:"我刚拿到Pendo,需要在我们的React SaaS应用上安装它。从哪里开始?"
技能处理:涵盖安装脚本代码段的放置位置、访客ID和账户ID策略(使用内部用户ID,而非邮箱)、元数据架构设计(角色、方案、注册日期)、初始页面和功能标记计划,以及首个引导的创建。标记常见错误:使用邮箱作为访客ID(邮箱变更时会失效)、未在UI中预定义元数据字段、第一天就过度标记功能。
结果:包含安装脚本、元数据架构和标记策略的完整设置计划。
Troubleshooting
故障排除
Guides not appearing for target users
目标用户未看到引导
Symptom: Guide is active but users don't see it.
Cause: CSS selector changed after a deploy, segment targeting is too narrow, or display frequency limit reached.
Solution: 1) Check the guide's element targeting — if using auto-detected selectors, switch to a attribute. 2) Preview the guide on your own account to verify targeting. 3) Check display rules — "show once" means users who dismissed it won't see it again. 4) Verify the user matches the guide's segment.
data-pendo症状:引导已激活,但用户看不到。
原因:部署后CSS选择器变更、受众细分定位过窄,或达到展示频率限制。
解决方案:1) 检查引导的元素定位——如果使用自动检测的选择器,切换为属性。2) 在自己的账户上预览引导,验证定位是否正确。3) 检查展示规则——"仅展示一次"意味着关闭过引导的用户不会再次看到。4) 验证用户是否符合引导的受众细分条件。
data-pendoAnalytics showing data discrepancies
分析数据存在差异
Symptom: Feature usage numbers in Pendo don't match your own analytics.
Cause: Different event definitions, Pendo deduplication logic, or install script not firing on all pages.
Solution: 1) Verify the install script loads on every page (check for SPA route-change handling). 2) Compare event definitions — Pendo counts unique visitors per day by default, not raw events. 3) Check if ad blockers are suppressing the Pendo agent in your user base. 4) Use the browser console () to verify the agent is running.
pendo.validateInstall()症状:Pendo中的功能使用数据与你自己的分析数据不匹配。
原因:事件定义不同、Pendo去重逻辑不同,或安装脚本未在所有页面加载。
解决方案:1) 验证安装脚本是否在每个页面加载(检查SPA路由变更的处理)。2) 对比事件定义——Pendo默认按日统计独立访客数,而非原始事件数。3) 检查用户群体中是否有广告拦截器阻止了Pendo代理。4) 使用浏览器控制台()验证代理是否在运行。
pendo.validateInstall()Feature/page tagging is overwhelming
功能/页面标记工作量过大
Symptom: Hundreds of untagged pages and features, unclear where to start.
Cause: Pendo captures all clicks and page views but requires manual tagging to make them meaningful.
Solution: 1) Start with your top 5-10 features by business value — don't try to tag everything. 2) Use URL rules for pages (wildcards for dynamic segments). 3) For features, tag the primary CTA element, not every clickable item. 4) Establish a naming convention early (Module > Feature > Action). 5) Assign tagging ownership by product area.
症状:数百个未标记的页面和功能,不清楚从何处入手。
原因:Pendo会捕获所有点击和页面浏览,但需要手动标记才能使其有意义。
解决方案:1) 从业务价值最高的5-10个功能开始——不要试图标记所有内容。2) 对页面使用URL规则(动态段使用通配符)。3) 对于功能,标记主要的CTA元素,而非每个可点击项。4) 尽早建立命名规范(模块>功能>操作)。5) 按产品领域分配标记负责人。