# PostHog custom analytics  that build themselves.

Connect your PostHog instance and build custom dashboards, feature flag management tools, and experiment result viewers — just by describing what you need.

## Feature Flags — ProductionSynced

| Flag                     | Status   | Rollout |  
|--------------------------|----------|---------|  
| New onboarding flow      | Rolling out | 45%     |  
| AI-powered search        | Testing  | 10%     |  
| Dark mode redesign       | Shipped  | 100%    |  
| Collaborative editing     | Testing  | 5%      |  
| Usage-based billing       | Rolling out | 30%    |

## Ship features with confidence.

14  
feature flags managed  
3.2M  
events analyzed  
92%  
experiment win rate  
<5min  
to custom dashboard

## See where users drop off. Build what brings them back.

Feature Adoption — new-onboarding-v2

| Step               | Users   | Drop-off |  
|---------------------|---------|----------|  
| Feature released     | 12,847  | 100%     |  
| Discovered           | 8,736   | 68%      |  
| First used          | 5,267   | 41%      |  
| Adopted (7d+)       | 3,597   | 28%      |  
| Power user          | 1,542   | 12%      |

## A/B tests that explain themselves.

Build experiment result portals your whole team can understand. Confidence intervals, winner declarations, no statistics degree required.

### Experiment Results

#### Checkout Flow Redesign

Simplifying the checkout to 2 steps will increase conversion by 15%+

- Significant  
- Started Feb 12, 2026  
- 21 days  
- 48,291 participants  
- p = 0.0013

**Control**  
- 3.2% conversion rate  
- n = 24,146

**Winner**  
- Variant A  
- 4.8% conversion rate  
- n = 24,145  
- Relative lift +50%  
- Confidence 97.3%

Statistical significance 95% threshold

## What product engineers build with camelAI.

### Feature adoption dashboards

Track every feature flag rollout from release to power-user adoption. See adoption curves, retention by cohort, and the features that actually stick — all in a shareable dashboard your PM will love.

### Experiment result portals

Build an A/B test results viewer for your whole team. Confidence intervals, sample sizes, conversion lifts, and clear winner declarations — no statistics degree required.

### User journey analyzers

Funnels, user paths, and drop-off analysis. Ask questions like “where do users churn after signup?” and get interactive visualizations you can drill into.

### Automated metric alerts

Set up cron jobs that query PostHog on a schedule, check for anomalies in your product metrics, and ping you in Slack when something needs attention.

## We love open source too.

PostHog is open source. Your camelAI apps publish to the open web. No lock-in, no walled gardens — just your data, your tools, your way.

### Open data

Your PostHog instance, your events, your queries. camelAI reads your data — never stores, never locks you in.

### Open deployment

Publish to *.camelai.app and share with anyone. No walled garden. Your dashboards live on the open web.

### Open integration

50+ connections beyond PostHog. Pull in Stripe revenue, Slack alerts, GitHub deploys — one unified view.

```typescript
const posthog = await connect("posthog");

// Query events, feature flags, experiments...

const funnel = await posthog.query({
  events: ["signup","onboarding_complete","first_project"],
  type: "funnel",
  dateRange: "last_30_days"
});
```

## What will you build on your PostHog data?

- “Connect to our PostHog instance and build a feature adoption dashboard. Show rollout percentage, daily active users per flag, and a retention curve for each feature released this quarter.”

- “Pull our A/B experiment results from PostHog. Build a portal where anyone on the team can see conversion rates, confidence intervals, and which variant won — with filters by date and experiment name.”

- “Analyze our signup funnel from PostHog events: landing_page → signup_start → email_verified → first_project → invited_teammate. Show drop-off percentages and breakdown by acquisition source.”

- “Set up a cron job that checks our PostHog metrics every morning. If daily active users drops more than 15% week-over-week, post an alert to #product-alerts in Slack with a breakdown by platform.”

## Your product data is already there.

Now make it actionable.