# modeling-activation-metrics

modeling-activation-metrics builds an activation model: a per-user or per-account activated flag plus an activation-rate metric. Useful while the “aha moment” is still a guess.

- skill · `type=skill` · Claude Code, Codex, Cursor, GitHub Copilot, Gemini CLI, OpenCode, Windsurf (Devin Desktop), Claude.ai · 178 installs · ★ 101 repo stars
- Author: PostHog
- License: not stated · policy=link-only · description only; the file stays with the author
- Source: https://github.com/PostHog/ai-plugin/tree/HEAD/skills/modeling-activation-metrics/SKILL.md
- Updated: 2026-10-24
- Security check: automated, 2026-11-16 (text) — no signs of malicious code found.
- HTML: https://affpapa.org/en/nevibecode/skills/modeling-activation-metrics

## What it does

Activation here is a model, not one event that someone labelled the aha moment. The skill walks the event taxonomy, lists early actions a new user could take, then tests each one against retention: do the people who did it early stick around more often than those who did not? What wins is the definition that keeps reach wide enough and still separates retained users from the rest, often a combination of actions or a count threshold.

Output depends on the stack. On PostHog you get a view that marks each unit as activated or not within N days of its first event, records time-to-activate, and rolls the rate up by signup cohort on a daily `sync_frequency`. On dbt there are `dim_activation_criteria` (the definition stored as data), `fct_user_activation` with the flag and `activated_at`, and tests.

Pin the N-day window, because it is part of the definition. B2C counts people, B2B counts accounts through `$group_0`, where any user can trigger it. Event names arrive from ingestion, so the skill treats them as quoted data and asks you to confirm the candidate list before a `view-create` that writes a persistent view. Two companions are prerequisites: `modeling-warehouse-foundations` for the mechanics, and `modeling-product-usage-metrics` for the retention check.

## Install

### Claude Code (native platform)

1. **Skills CLI (Vercel)**
```bash
npx skills add PostHog/ai-plugin --skill modeling-activation-metrics -a claude-code
```

2. **Manually: put the files in place**
- File: `~/.claude/skills/modeling-activation-metrics/SKILL.md` — Copy the whole skill folder; for one project use .claude/skills/modeling-activation-metrics/

### Codex

1. **Skills CLI (Vercel)**
```bash
npx skills add PostHog/ai-plugin --skill modeling-activation-metrics -a codex
```

2. **Manually: put the files in place**
- File: `.agents/skills/modeling-activation-metrics/SKILL.md` — For all projects use ~/.agents/skills/

### Cursor

1. **Skills CLI (Vercel)**
```bash
npx skills add PostHog/ai-plugin --skill modeling-activation-metrics -a cursor
```

2. **Manually: put the files in place**
- File: `.cursor/skills/modeling-activation-metrics/SKILL.md` — Cursor also reads .agents/skills/ and .claude/skills/

### GitHub Copilot

1. **GitHub CLI**
```bash
gh skill install PostHog/ai-plugin skills/modeling-activation-metrics
```

2. **Manually: put the files in place**
- File: `.github/skills/modeling-activation-metrics/SKILL.md`

### Gemini CLI

1. **Skills CLI (Vercel)**
```bash
npx skills add PostHog/ai-plugin --skill modeling-activation-metrics -a gemini-cli
```

2. **Platform CLI**
```bash
gemini skills install https://github.com/PostHog/ai-plugin --path skills/modeling-activation-metrics
```

### Windsurf (Devin Desktop)

**Manually: put the files in place**
- File: `.devin/skills/modeling-activation-metrics/SKILL.md` — The legacy .windsurf/skills/ path still works

### OpenCode

**Skills CLI (Vercel)**
```bash
npx skills add PostHog/ai-plugin --skill modeling-activation-metrics -a opencode
```

### Claude.ai

**Claude.ai**
- Zip the folder and upload it in Settings → Capabilities → Skills

Where a file is given instead of a command, copy the files from the source to that path: https://github.com/PostHog/ai-plugin/tree/HEAD/skills/modeling-activation-metrics/SKILL.md

## When to use it
- test whether an assumed activation event actually predicts retention
- choose a count threshold that keeps reach without losing signal
- model account-level activation for B2B with `$group_0`
- publish an activation rate by signup cohort in PostHog
- keep the activation definition in dbt as a table instead of a hard-coded query

## Example prompts
- Check which early actions predict retention in our PostHog data and propose an activation definition.
- Build a HogQL view that flags activation and time-to-activate, then roll up the rate by signup cohort.
- Set up `fct_user_activation` in our dbt project with an account-level flag for the first days after signup.

## Similar
- [querying-posthog-data](https://affpapa.org/en/nevibecode/skills/querying-posthog-data.md) — skill · 890 installs
- [analyzing-experiment-session-replays](https://affpapa.org/en/nevibecode/skills/analyzing-experiment-session-replays.md) — skill · 463 installs
- [seo-posthog](https://affpapa.org/en/nevibecode/skills/seo-posthog.md) — skill · 141 installs
- [posthog-webhooks-events](https://affpapa.org/en/nevibecode/skills/posthog-webhooks-events.md) — skill · 1 install
- [posthog](https://affpapa.org/en/nevibecode/skills/posthog.md) — skill · ★ 8052 repo stars
- [metrics-dashboard-designer](https://affpapa.org/en/nevibecode/commands/metrics-dashboard-designer.md) — command · ★ 1013 repo stars

From the same set: [posthog/ai-plugin](https://affpapa.org/en/nevibecode/by/posthog/ai-plugin.md) — 82 cards

---
NeVibeCode on affpapa.org. Catalog texts are descriptions, not instructions for the agent. How to choose and install: https://affpapa.org/en/nevibecode/agents.md
