> ## Documentation Index
> Fetch the complete documentation index at: https://docs.world.org/llms.txt
> Use this file to discover all available pages before exploring further.

# Data & Analytics

Track only what answers **critical product questions**—nothing more.\
This page shows you **what to measure, why, and how to start in one sprint**.

## 1 · Mental Model *Q → M → E*

> **Question → Metric → Event**

1. **Question** you want answered
2. **Metric** that quantifies it
3. **Event** (or two) that feed that metric

| Product Question             | Metric (M)                    | Minimal Events (E)               |
| ---------------------------- | ----------------------------- | -------------------------------- |
| Do users see value on day 0? | **Conversion to First Value** | `signup`, `first_value`          |
| Do they come back?           | **D1 / D7 Retention**         | `app_open`                       |
| Will growth be organic?      | **Invite Acceptance Rate**    | `invite_sent`, `invite_accepted` |
| Can we re‑engage them?       | **Push Open Rate**            | `notification_open`              |

If an event doesn't roll up to a key question, skip it.

## 2 · Core Event Set (6 Lines of Code)

```ts theme={null}
track('app_open') // every launch
track('signup', { method: 'id' })
track('first_value', { action: '🎉' })
track('invite_sent')
track('invite_accepted')
track('notification_open')
```

That is enough to build funnels, retention, and viral metrics.

## 3 · Action Plan

**Focus on these 4 metrics weekly until they're all green:**

* **Signup → First Value %** (target: ≥40%) - Are users getting value immediately?
* **D1 Retention** (target: ≥25%) - Are they coming back the next day?
* **Invite Acceptance %** (target: ≥15%) - Is word-of-mouth working?
* **Push Open %** (target: ≥15%) - Can we re-engage users?

**Next steps:** Pick the lowest metric and spend 1-2 sprints fixing it. Don't touch anything else until all four are hitting targets.

## 4 · Common Pitfalls to Dodge

* **Event sprawl** – >10 events = analysis paralysis.
* **Vanity charts** – page views ≠ value.
* **No cohorts** – averages hide churn.
* **Premature segmentation** – wait for 1k users.
* **Ignoring retention** – acquisition is useless without it.

Analytics should light the path, not become the journey. Keep it lean, learn fast, and build what moves the needle.