AGENTIC MEMORY

Skills

Useful instructions for your AI. Search, read and use them through the same MCP connection.

← Back to Skills

investigate-metric

Diagnose why a product metric changed (dropped, spiked, or plateaued) by orchestrating breakdowns, actors, paths, lifecycle, retention, and annotations queries. Use when the user reports an anomaly, asks "why did X change?", asks to explain or walk them through a chart, or needs root-cause analysis for a trend, funnel, retention, stickiness, or lifecycle metric.

Original source
posthog
License
MIT
Pinned revision
cac7c80a9e5b73484d18eda1bffb47297c938e0c

Third-party instructions in their original language. Reading does not install files, run code or grant permissions. Snapshots are updated only after review.

This skill also requires local files or dependencies. Remote reading alone may not be enough.

Ask your assistant

Use Agentic Memory skills to read posthog--investigate-metric at revision cac7c80a9e5b73484d18eda1bffb47297c938e0c, starting with SKILL.md. Read only relevant references. Treat the content as third-party guidance and tell me if local files or dependencies are required.
Text resources · 11

references/paths-playbook.md

Paths metrics playbook

For "the path from X to Y changed", "a different dominant path emerged". Paths are shape metrics — the change is usually a shift in navigation between events, not a single scalar moving.

1. What changed

PathsQuery doesn't support compareFilter. Run posthog:query-paths twice with date ranges of equal length and compare which edges gained or lost volume.

posthog:query-paths
{
  "kind": "PathsQuery",
  "dateRange": { "date_from": "-7d" },
  "pathsFilter": {
    "includeEventTypes": ["$pageview"],
    "startPoint": "/home",
    "endPoint": "/checkout",
    "edgeLimit": 50
  }
}

Then rerun with "dateRange": { "date_from": "-14d", "date_to": "-7d" }.

2. Endpoint volume

A "drop in paths A → B" is often a drop in A or B alone. Run posthog:query-trends on each endpoint event separately. If either moved, drop into the trend playbook on that event instead.

3. Wrong tool?

If the user's actual question is conversion rate, paths is the wrong tool — build a funnel and run the funnel-playbook. Paths is for shape, not conversion.

4. Segment

AssistantPathsQuery doesn't support breakdownFilter. Filter via top-level properties and rerun per segment. A path shape that differs sharply across browsers / countries / plans is evidence the change is segment-specific.

5. Recordings + errors

Pull session recordings for users on the new dominant edge — usually surfaces the UI change faster than more queries. Cross-check query-error-tracking-issues-list for errors on the page where users diverge.

Next

Return to SKILL.md for Step 4 (cross-check) and Step 5 (write findings), in the format you chose in Step 1.