Back to Details

launch-metrics

Compare original and translation side by side

🇺🇸

Original

English
🇨🇳

Translation

Chinese

Launch Metrics

发布指标

This skill covers how to measure product launches — what to set up before launch day, what to track during, and how to evaluate results afterward. Without consistent measurement, launches are guesswork. With it, each launch informs the next.
此 skill 介绍如何衡量产品发布——发布日前需要做哪些准备、发布期间需要追踪哪些数据，以及发布后如何评估结果。没有一致的衡量方式，发布就只能靠猜测；建立衡量机制后，每次发布都能为下一次提供参考。

Pre-launch setup

发布前设置

Measurement starts before launch day. Setting up tracking the day of the launch means you miss the first-wave data when traffic is highest.
衡量工作应从发布日前开始。等到发布当天再设置追踪，就会错过流量最高时的第一波数据。

What to set up

需要设置的内容

  1. UTM parameters — Tag every link in every distribution channel
  2. Analytics events — Instrument the conversion funnel end-to-end
  3. Baseline capture — Record current metrics before launch so you can measure lift
  4. Dashboard or view — One place to watch all key metrics on launch day
  1. UTM 参数——为每个分发渠道中的每个链接添加标签
  2. 分析事件——对整个转化漏斗进行埋点
  3. 基线数据采集——记录发布前的现有指标，以便衡量提升幅度
  4. Dashboard 或视图——集中查看发布当天的所有关键指标

UTM conventions

UTM 规范

UTM parameters track where traffic comes from. Use a consistent naming scheme across all launches so you can compare over time.
utm_source    = the platform (twitter, linkedin, email, producthunt, hackernews)
utm_medium    = the channel type (social, email, paid, referral)
utm_campaign  = the launch name (feature-name-launch, v2-launch, etc.)
utm_content   = the specific asset (tweet-thread, announcement-email, blog-post)
Example URLs:
  • Twitter announcement: 
    ?utm_source=twitter&utm_medium=social&utm_campaign=realtime-api-launch&utm_content=launch-tweet
  • Launch email: 
    ?utm_source=email&utm_medium=email&utm_campaign=realtime-api-launch&utm_content=announcement-email
  • Product Hunt: 
    ?utm_source=producthunt&utm_medium=referral&utm_campaign=realtime-api-launch&utm_content=ph-listing
Rules:
  • Always lowercase
  • Use hyphens, not underscores or spaces, in values
  • Campaign name should match across all channels for a single launch
  • Create all UTMs from a spreadsheet before launch day so nothing is missing
UTM 参数用于追踪流量来源。所有发布应采用一致的命名方案，以便进行跨期比较。
utm_source    = the platform (twitter, linkedin, email, producthunt, hackernews)
utm_medium    = the channel type (social, email, paid, referral)
utm_campaign  = the launch name (feature-name-launch, v2-launch, etc.)
utm_content   = the specific asset (tweet-thread, announcement-email, blog-post)
URL 示例：
  • Twitter 公告：
    ?utm_source=twitter&utm_medium=social&utm_campaign=realtime-api-launch&utm_content=launch-tweet
  • 发布邮件：
    ?utm_source=email&utm_medium=email&utm_campaign=realtime-api-launch&utm_content=announcement-email
  • Product Hunt：
    ?utm_source=producthunt&utm_medium=referral&utm_campaign=realtime-api-launch&utm_content=ph-listing
规则：
  • 始终使用小写字母
  • 参数值中使用连字符，不要使用下划线或空格
  • 同一次发布的所有渠道应使用相同的 Campaign 名称
  • 在发布日前使用电子表格创建所有 UTM，确保没有遗漏

Instrumented events

埋点事件

Before launch, verify these analytics events exist and fire correctly:
EventDescription
landing_page_view
Landing page loaded with UTM params captured
signup_started
User begins registration
signup_completed
Account created
first_action
First meaningful product action (varies by product)
cta_click
Each CTA button clicked (with label)
blog_post_view
Launch blog post loaded
email_open
Launch email opened (from email platform)
email_click
Link clicked in launch email
If these events don't exist and you don't have time to add them before launch, at minimum use UTMs to measure traffic and signups from whatever analytics platform you already have.
发布前，验证以下分析事件已经存在并能正确触发：
事件说明
landing_page_view
落地页加载，并已捕获 UTM 参数
signup_started
用户开始注册
signup_completed
账户已创建
first_action
用户首次执行有实际意义的产品操作（因产品而异）
cta_click
用户点击 CTA 按钮（包含按钮标签）
blog_post_view
发布博客文章已加载
email_open
发布邮件已被打开（来自邮件平台）
email_click
发布邮件中的链接已被点击
如果这些事件尚不存在，而且来不及在发布前添加，至少应使用 UTM，并通过现有分析平台衡量流量和注册用户数。

KPIs by tier

各层级 KPI

What "success" looks like depends on launch tier. Don't compare a Tier 3 changelog tweet to a Tier 1 product launch.
不同发布层级的“成功”标准各不相同。不要将 Tier 3 更新日志推文与 Tier 1 产品发布进行比较。

Tier 1 KPIs (full product launch)

Tier 1 KPI（完整产品发布）

MetricWhat it measuresWhere to track
New signups (day 1, week 1)Direct acquisition impactProduct analytics
Landing page visitorsReach of the announcementWeb analytics
Landing page → signup conversion ratePage effectivenessWeb analytics
Email open rateList engagementEmail platform
Email click rateEmail copy effectivenessEmail platform
Social impressions + engagementsAwarenessNative platform analytics
Press mentionsPR reachManual + Google Alerts
Backlinks generatedSEO impactAhrefs, Search Console
Product Hunt rank (if applicable)PH-specific reachProduct Hunt
指标衡量内容追踪位置
新注册用户数（第 1 天、第 1 周）直接获客影响产品分析
落地页访问者公告触达范围Web 分析
落地页 → 注册转化率页面 effectivenessWeb 分析
邮件打开率邮件列表互动度邮件平台
邮件点击率邮件文案效果邮件平台
社交内容曝光量 + 互动量认知度各平台原生分析工具
媒体报道提及PR 触达人工统计 + Google Alerts
生成的 backlinks 数量SEO 影响Ahrefs、Search Console
Product Hunt 排名（如适用）Product Hunt 特定渠道触达Product Hunt

Tier 2 KPIs (major feature)

Tier 2 KPI（主要功能发布）

MetricWhat it measures
New signups from launch trafficAcquisition
Feature adoption rate (existing users)Activation
Blog post viewsContent reach
Social engagement rateResonance
Email open + click rateList engagement
指标衡量内容
来自发布流量的新注册用户数获客
功能采用率（现有用户）激活
博客文章浏览量内容触达
社交互动率共鸣度
邮件打开率 + 点击率邮件列表互动度

Tier 3 KPIs (small improvement)

Tier 3 KPI（小改进）

MetricWhat it measures
Changelog viewsAwareness among existing users
Social post engagementReach
指标衡量内容
更新日志浏览量现有用户中的认知度
社交帖子互动量触达

Benchmark expectations

基准预期

These are rough benchmarks for a B2B/developer tool with an established audience. Use them to calibrate expectations, not as hard targets.
MetricWeakSolidStrong
Email open rate (existing list)< 20%30-40%> 50%
Email click rate< 2%4-8%> 10%
Landing page → signup CVR< 2%5-10%> 15%
Twitter launch tweet impressions (10K followers)< 5K10-30K> 50K
HN front page (if hit)—200-500 visits2,000+ visits
Product Hunt (featured)< 100 upvotes200-500500+
Tier 1 launch: week-1 signups< 100200-1,0001,000+
These vary enormously by audience size, product category, and launch quality. Treat them as directional.
以下基准适用于拥有稳定受众的 B2B/开发者工具。这些数据用于校准预期，而不是硬性目标。
指标较弱稳健强劲
邮件打开率（现有邮件列表）< 20%30-40%> 50%
邮件点击率< 2%4-8%> 10%
落地页 → 注册 CVR< 2%5-10%> 15%
Twitter 发布推文曝光量（1 万关注者）< 5K10-30K> 50K
HN 首页（如登上）—200-500 次访问2,000+ 次访问
Product Hunt（featured）< 100 个投票200-500500+
Tier 1 发布：第 1 周注册用户数< 100200-1,0001,000+
这些数据会因受众规模、产品类别和发布质量而产生巨大差异，应将其视为方向性参考。

Launch day monitoring

发布日监控

On launch day, watch metrics in near real-time for the first 4-6 hours. This is when you can still adjust (reshare a post, send a follow-up tweet, extend a limited offer).
What to watch:
  • Signups per hour (alert if significantly below expectation)
  • Top traffic sources by signup conversion rate (not just volume)
  • Email delivery issues (bounce rate, spam complaints)
  • Social engagement velocity in the first 2 hours (predicts final reach)
Don't do:
  • Obsessively refresh metrics while neglecting to engage with social comments
  • Make major messaging changes mid-launch based on early data
  • Panic if a specific channel underperforms — wait for 24-hour data
发布当天，应在最初 4-6 小时内近实时关注指标。这段时间仍可进行调整，例如重新分享帖子、发送后续推文或延长限时优惠。
需要关注：
  • 每小时注册用户数（如果明显低于预期则发出警报）
  • 按注册转化率排名的主要流量来源（不要只看流量总量）
  • 邮件送达问题（退信率、垃圾邮件投诉）
  • 最初 2 小时内的社交互动速度（可预测最终触达）
不要做：
  • 反复刷新指标，却忽视与社交评论互动
  • 根据早期数据，在发布中途大幅调整信息传达策略
  • 因某个特定渠道表现不佳而惊慌；应等待完整的 24 小时数据

Post-launch retro

发布后复盘

Run the retro within 5-7 days of launch, when the data is fresh. For Tier 1, schedule this during launch prep. For Tier 2, a Slack thread async works.
应在发布后 5-7 天内进行复盘，以趁数据仍然新鲜。对于 Tier 1，应在发布准备阶段安排好时间；对于 Tier 2，使用异步 Slack 话题串即可。

Retro data to pull

复盘前需要提取的数据

Before the retro, gather:
  • Traffic by source (UTM breakdown)
  • Signups by source
  • Conversion rates by source (signups ÷ visitors)
  • Email metrics (open, click, unsubscribe)
  • Social metrics by platform
  • Comparison to previous comparable launch (if available)
复盘前应收集：
  • 各来源的流量（UTM 分解）
  • 各来源的注册用户数
  • 各来源的转化率（注册用户数 ÷ 访问者数）
  • 邮件指标（打开、点击、退订）
  • 各平台的社交指标
  • 与上一次同类发布的数据对比（如有）

Questions to answer

需要回答的问题

  1. What was the top-performing channel (by conversion rate, not just volume)?
  2. What was the worst-performing channel relative to effort invested?
  3. What was surprising? (channels that over- or under-performed expectations)
  4. Did messaging land? (qualitative: comments, replies, email responses)
  5. Where did the funnel leak? (high traffic, low signups → landing page problem; low traffic → distribution problem)
  6. What would you do differently?
  1. 哪个渠道表现最佳（按转化率判断，而不只是流量总量）？
  2. 相对于投入力度，哪个渠道表现最差？
  3. 有哪些意外情况（哪些渠道的表现超出或低于预期）？
  4. 信息传达是否有效（定性分析：评论、回复、邮件反馈）？
  5. 漏斗在哪个环节流失（流量高但注册少 → 落地页存在问题；流量低 → 分发存在问题）？
  6. 下次会做出哪些不同调整？

Retro output

复盘输出

The retro should produce exactly two things:
  1. Updated distribution priorities — which channels to invest more/less in next time
  2. Checklist updates — additions or removals to the launch checklist based on what you learned
Document in a shared place (Notion, Linear, Confluence) so it accumulates over launches. The value compounds: after 10 launches, you have a quantitative model of what works for your product and audience.
每次复盘应恰好产出以下两项内容：
  1. 更新后的分发优先级——确定下次应增加或减少哪些渠道的投入
  2. 更新后的检查清单——根据新发现增加或删除发布检查清单中的项目
将结果记录在共享位置（Notion、Linear、Confluence），使其能够随每次发布不断积累。价值会不断累积：完成 10 次发布后，就能建立一套关于哪些方法适合自身产品和受众的量化模型。

Building a launch history

构建发布历史

Track every launch in a single spreadsheet or database:
LaunchDateTierTop channelSignups (D1)Signups (W1)Email open rateNotes
After 5+ launches, patterns emerge:
  • Which channels consistently convert for your audience
  • What tier of launch drives what magnitude of signups
  • How your benchmarks trend over time (growth in list size, social following)
This history is your most valuable launch asset. It makes future launch planning faster and more accurate than any external benchmark.
使用单个电子表格或数据库追踪每次发布：
发布日期层级最佳渠道注册用户数（D1）注册用户数（W1）邮件打开率备注
完成 5 次以上发布后，规律便会逐渐显现：
  • 哪些渠道始终能为你的受众带来更高转化
  • 不同发布层级分别能带来多少注册用户
  • 各项基准如何随时间变化（邮件列表规模、社交关注者数量）
这份发布历史是最有价值的发布资产。它能比任何外部基准都更快、更准确地指导未来的发布规划。

Companion skills

配套 skill

  • launch-strategy — defines the launch tier and checklist that determines what to track
  • launch-distribution — each channel's UTM is set up before launch day
  • launch-email — email open/click metrics feed back into the retro
  • launch-blog-post — blog post UTMs and organic search performance over time
  • launch-strategy——定义发布层级和检查清单，用于决定需要追踪哪些指标
  • launch-distribution——在发布日前为每个渠道设置 UTM
  • launch-email——将邮件打开率和点击率纳入复盘
  • launch-blog-post——持续追踪博客文章的 UTM 和自然搜索表现