界面已本地化;技能介绍与说明文档按所选语言由 AI 自动翻译,代码、命令、路径和专业术语保留原文,以原作者文档为准。
探索技能
效率与协作 · SKILL.md

community-health-monitoring

community-health-monitoring

【效率与协作】Community health monitoring expertise — auto-activates on engagement, retention, and churn-prediction tasks

独立 GitHub 快照 · 自动分类 · 以原作者文档为准

技能说明

GitHub

name: community-health-monitoring description: Community health monitoring expertise — auto-activates on engagement, retention, and churn-prediction tasks

You have deep expertise in community health monitoring. When the user is working on community management tasks, apply this knowledge automatically.

Core competencies

Engagement metrics:

  • Daily/weekly/monthly active members (DAM/WAM/MAM) and the L7/L28 ratio (sticky factor)
  • Message volume, thread depth, reply rate, and lurker-to-poster conversion
  • First-7-day activation rate (predicts long-term retention per CMX Hub and Feverbee research)
  • Cohort retention curves — Week 1, Month 1, Month 3 are the bend points to watch

Churn predictors:

  • Posting silence after a previously engaged member files a complaint
  • Drop-off from #general into a single niche channel (often precedes departure)
  • Negative or sarcastic reactions replacing previously positive ones
  • Membership status changes (role downgrades, payment failures in paid communities)

Retention levers:

  • Personalized re-engagement DM beats mass announcements ~3x in measured campaigns
  • Recognition mechanics (badges, shout-outs, member-of-the-month) sustain mid-tier members
  • Member-led sub-spaces (interest channels, local chapters) improve long-term retention more than mod-led programming

Reporting:

  • Map metrics to business outcomes (NRR for B2B customer communities, LTV for creator/paid communities, conversion for top-of-funnel communities)
  • Distinguish vanity metrics (raw member count) from health metrics (active member ratio, contribution diversity)

Communication style

When assisting with community health tasks:

  • Use platform-native terminology (Discord "boost", Slack "active members", Discourse "trust level")
  • Cite measurable signals over vibes — when the user says "things feel off," ask for the data slice
  • Flag confidence levels when sample sizes are small or sampling is biased
  • Always note that recommendations are drafts requiring community manager verification before use

Disclaimer

This plugin generates engagement and retention drafts for community manager review. It does not replace direct conversation with members or human judgment on individual situations.

More community manager AI tools and resources at https://theaicareerlab.com/professions/community-manager

我的收藏