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

# Analytics and Exports

> Choose the right downstream pattern for BI, QA, and operational dashboards.

# Choose the right retrieval pattern for the job.

Use webhooks when you need low-latency event handling. Use call-log exports when analysts need wide time windows. Use direct API pulls for near-real-time dashboards that can tolerate polling.

<CardGroup cols={3}>
  <Card title="Webhooks" icon="zap">Best for event-driven automation and near-real-time workflows.</Card>
  <Card title="Direct API" icon="database">Best for dashboards and controlled reconciliation jobs.</Card>
  <Card title="Exports" icon="sheet">Best for warehouse loads, finance review, and historical QA batches.</Card>
</CardGroup>

## Decision matrix

| Need                     | Recommended pattern | Notes                                                       |
| ------------------------ | ------------------- | ----------------------------------------------------------- |
| Real-time trigger        | Webhooks            | Build idempotent consumers with replay tolerance            |
| Hourly dashboard refresh | Direct API          | Use cursor/page semantics correctly and checkpoint progress |
| Historical backfill      | Export jobs         | Prefer async exports for wide date windows                  |

## Reliable data pipeline workflow

1. Capture real-time outcomes via webhooks.
2. Run scheduled API reconciliation for missed events.
3. Schedule weekly/monthly exports for audit and finance.
4. Deduplicate in warehouse by stable session/contact identifiers.

<Note>
  For large datasets, mixing webhook ingestion and export reconciliation gives better correctness than either pattern alone.
</Note>