Firehose is Everflow’s real-time event streaming service. Instead of polling the REST API for new data, Firehose pushes events to your infrastructure as they happen — with sub-second latency and no rate limiting.Documentation Index
Fetch the complete documentation index at: https://developers.everflow.io/llms.txt
Use this file to discover all available pages before exploring further.
When to use Firehose vs streaming vs export
Everflow gives you three ways to move data out of the platform at scale. Pick by use case:| Firehose | Streaming endpoints (/clicks/stream, /conversions/stream) | Export endpoints (/conversions/export, etc.) | |
|---|---|---|---|
| Delivery model | Pushed to you in real time | You pull, server streams the response | You pull, server returns a file (CSV/JSON) |
| Latency | Sub-second | Seconds, depending on volume | Minutes for large windows |
| Volume per request | Unbounded — every event flows continuously | Bounded by date range | Bounded by date range; capped by file size |
| Rate limits | None | Counts against the API rate limit | Counts against the API rate limit |
| Use when | You want a live event pipeline (warehouse, dashboards, automation) | You’re backfilling or pulling a fixed historical window | You need a downloadable file for ad-hoc analysis or third-party tools |
| Setup | Provisioned by Everflow Support — see below | Available out of the box | Available out of the box |
Supported event types
Firehose can stream any combination of these event types:| Event type | Description |
|---|---|
| Clicks | Fired when a user clicks a tracking link. Includes device info, geolocation, query parameters, and attribution data. |
| Impressions | Fired when an impression pixel loads. Includes device and geolocation data. Requires the impression tracking package. |
| Conversions | Fired when a conversion event is recorded. Includes conversion details, payout/revenue, event names, and custom parameters. |
| Conversion updates | Fired when a conversion status changes (e.g., approved, rejected) or when payout/revenue values are modified. |
Supported delivery platforms
Firehose supports the following delivery methods:Amazon SQS
Events are delivered to an Amazon Simple Queue Service (SQS) queue. Supports both Standard and FIFO queue types.- Standard queues — Best-effort ordering, at-least-once delivery, maximum throughput
- FIFO queues — Guaranteed ordering within message groups, exactly-once processing
- AWS Region
- SQS Queue URL
- IAM credentials or cross-account role ARN for Everflow to publish to your queue
Google Cloud Pub/Sub
Events are published to a Google Cloud Pub/Sub topic. Supports optional message attributes for custom routing. You provide:- Google Cloud Project ID
- Pub/Sub Topic name
- Service account credentials for Everflow to publish to your topic
HTTP/HTTPS POST
Events are delivered as JSON payloads via HTTP POST requests to your custom endpoint. You provide:- The URL to receive events
- Any required authentication headers (e.g., API keys, bearer tokens)
Azure Event Hubs
Available on request. Contact Everflow Support to configure Azure Event Hubs delivery.Event payloads
Each event is delivered as a JSON object. The payload structure is configurable — you can choose which fields to include and define custom field mappings.Click payload example
Impression payload example
Conversion payload example
Conversion update payload example
The conversion update payload is identical to the conversion payload, with the addition of anupdate_timestamp field indicating when the update occurred.
Payload configuration
When setting up your Firehose, you can customize:- Field selection — Choose which fields to include in each event type
- Field mapping — Rename fields to match your data schema (e.g., rename
affiliate_idtopartner_id) - Raw query parameters — Optionally include the full raw query string from the tracking URL
- Event filtering — Stream only specific event types (e.g., conversions only)
Setup
Firehose is configured by the Everflow team. To get started:- Choose your delivery platform — SQS, Pub/Sub, HTTP, or Azure
- Provision your infrastructure — Create the queue, topic, or endpoint on your side
- Contact Everflow Support — Provide your infrastructure details (queue URL, topic name, endpoint URL, and credentials)
- Configure your payload — Work with the Everflow team to define which fields and event types to include
- Start receiving events — Everflow enables the stream and events begin flowing in real time
Contact Support
Reach out to the Everflow Support team to set up Firehose for your account.
