signals-scout-customer-analytics

Detects per-account engagement cliffs, dormancy, and expansion signals in PostHog customer analytics.

713|118|Updated Aug 11, 2020
One-click install
npx skills add https://github.com/PostHog/posthog-foss --skill signals-scout-customer-analytics
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: signals-scout-customer-analytics
Source: https://github.com/PostHog/posthog-foss/tree/main/products/signals/skills/signals-scout-customer-analytics
Command: npx skills add https://github.com/PostHog/posthog-foss --skill signals-scout-customer-analytics

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Customer success and account teams cannot manually watch hundreds of accounts for early churn signals. This Skill proactively monitors per-account engagement in PostHog Customer Analytics, flagging accounts that are quietly disengaging or heating up before a renewal review forces the question.

Core Features & Use Cases

  • Churn-risk detection: Identifies engagement cliffs, dormancy onset, and champion departure (single-threading) on commercially-staked accounts by comparing each account against its own trailing baseline while the fleet holds steady.
  • Expansion signals: Surfaces accounts whose usage or active seats are climbing sharply versus baseline as upsell opportunities for account executives.
  • Durable watchlist memory: Maintains a scratchpad of account baselines, group-key join health, and dedupe pointers so coverage compounds across runs instead of restarting cold.
  • Use Case: A CSM asks which accounts are at risk this quarter. The scout scores staked accounts, finds one whose weekly events dropped from 4.2k to 1.1k while the fleet held steady, and files a P2 report with the account owner and evidence charts.

Quick Start

Ask the scout to check which of my commercially-staked accounts show declining engagement this week and report any churn risks.

Frequently Asked Questions about signals-scout-customer-analytics

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I detect churn risk in PostHog customer analytics?

Run this scout to score per-account engagement against each account's own trailing baseline. It flags accounts whose weekly events or active users drop sharply while the rest of the fleet holds steady, weighted by commercial ownership like assigned CSMs or CRM links.

How does the scout join accounts to event data?

It joins system.accounts to group-keyed events by matching each account's external_id against the discovered $group_N index. It first verifies overlap across group indexes, since rosters seeded from CRMs often fail to match live group keys.

What engagement patterns does the account health scout detect?

It detects engagement cliffs, dormancy onset, champion departure where a dominant user goes silent, and expansion signals where usage climbs sharply. Fleet-wide movements are excluded as they belong to other analytics scouts.

Why does the scout report no per-account engagement data?

This happens when the accounts roster's external_id values do not match any group key in the event stream, typically because the roster was seeded or CRM-imported. The scout records this as a join-unlinked config gap rather than filing false findings.

When does the scout skip reporting an account?

It skips accounts below the volume floor, new accounts without a baseline, unstaked accounts with no CRM link, known sandbox or seasonal accounts, and risks already covered by an existing inbox report, which it edits instead of duplicating.