> ## Documentation Index
> Fetch the complete documentation index at: https://mixpanel-edb78807-real-time-streams.mintlify.site/llms.txt
> Use this file to discover all available pages before exploring further.

# Real-Time Event Streams Overview

> Stream Mixpanel events to a custom webhook within seconds of ingestion to power activation, personalization, and downstream data workflows.

<Note>
  **Private beta.** Real-Time Event Streams are available to a limited set of customers during Phase 1. If you'd like to participate, reach out to your Mixpanel account team.
</Note>

Real-Time Event Streams deliver Mixpanel events to your infrastructure within seconds of ingestion, so you can react to user behavior without waiting for a batch export. Phase 1 ships a single destination type: a generic, Liquid-templated webhook. Native MarTech and AdTech integrations are planned for later phases based on beta feedback.

## What you get in the beta

<CardGroup cols={2}>
  <Card title="Low-latency delivery" icon="bolt">
    p95 delivery within 120 seconds of event ingestion.
  </Card>

  <Card title="Custom webhook destination" icon="webhook">
    Send events to any HTTPS endpoint you control, with your own auth and payload shape.
  </Card>

  <Card title="Event and property filters" icon="filter">
    Stream only the event types and property values that matter to your use case.
  </Card>

  <Card title="Liquid-templated payloads" icon="code">
    Fully control the outbound request body, including field names and identity.
  </Card>
</CardGroup>

## When to use streams

Use Real-Time Event Streams when latency matters and your workflow needs individual events as they happen:

* **Live activation** — trigger emails, pushes, or in-app messages seconds after a qualifying event.
* **Operational alerting** — feed on-call systems when key funnels break or error events spike.
* **Personalization** — update recommendations and experiences based on the user's current session.
* **Downstream pipelines** — fan events into your own queues or services for custom processing.

## How streams compare to Data Pipelines

[Data Pipelines](/docs/data-pipelines) deliver batches of events to warehouses and object stores on a schedule. Real-Time Event Streams deliver individual events with low latency to an HTTP endpoint you control. Use pipelines for analytics and reporting workloads; use streams when you need to act on events within seconds.

| Capability     | Data Pipelines             | Real-Time Event Streams               |
| -------------- | -------------------------- | ------------------------------------- |
| Delivery model | Batched exports            | Per-event HTTP delivery               |
| Latency        | Minutes to hours           | p95 within 120 seconds                |
| Destinations   | Warehouses, object storage | Custom webhook (Phase 1)              |
| Best for       | Analytics, reporting, BI   | Activation, alerting, personalization |

## Concepts

* **Destination** — the webhook endpoint and its authentication configuration.
* **Sync** — a rule that binds a set of filtered events and a Liquid payload template to a Destination.
* **Failed events log** — a 72-hour record of events that could not be delivered after all retries, retained with the destination HTTP response for debugging.

## Delivery guarantees at a glance

* **At-least-once delivery.** Every outbound request includes an `X-Mixpanel-Insert-Id` header so you can deduplicate on your side.
* **Automatic retries.** Transient failures (5xx, timeouts, network errors) are retried up to 5 times with exponential backoff over roughly 12 minutes. Permanent failures (4xx) are surfaced as configuration errors.
* **Latency SLA.** p95 ≤ 120 seconds from ingestion to a confirmed 2xx response at your endpoint under normal load.

See [Delivery and reliability](/docs/real-time-event-streams/delivery-and-reliability) for the full behavior.

## What's not in the beta

The following are intentionally out of scope for Phase 1 and will be evaluated for GA:

* Native destinations (Kafka, Kinesis, Pub/Sub, MarTech, AdTech).
* A field-mapping UI. Payload shaping is done entirely through the Liquid template.
* User-property backfill on historical events. Streams operate on newly ingested events only.

## Next steps

<CardGroup cols={2}>
  <Card title="Set up a webhook Sync" icon="gear" href="/docs/real-time-event-streams/setup">
    Configure a destination, choose events and property filters, and write your Liquid payload template.
  </Card>

  <Card title="Delivery and reliability" icon="shield-check" href="/docs/real-time-event-streams/delivery-and-reliability">
    Understand retries, the failed events log, and at-least-once delivery.
  </Card>

  <Card title="States and monitoring" icon="chart-line" href="/docs/real-time-event-streams/states-and-monitoring">
    Read Destination and Sync states and monitor Sync health.
  </Card>

  <Card title="Request beta access" icon="envelope" href="mailto:support@mixpanel.com">
    Reach out to your Mixpanel account team to join the private beta.
  </Card>
</CardGroup>