The Extractor in Datamorf is a specialized feature designed to perform reverse ETL (Extract → Transform → Load) operations. While normal workflows push data out of your systems (for example, sending CRM updates to Slack or creating records in a database), an Extractor does the opposite, it pulls data from external platforms like CRMs, spreadsheets, APIs, or databases and brings it into Datamorf automatically.

This makes the Extractor ideal for data synchronization, enrichment, and reporting automations where you need to periodically fetch data from another platform and process it inside Datamorf.

How the Extractor Works

Each Extractor follows a fixed three-step process that ensures consistency and reliability across all data sources:

1. Schedule — When It Runs

You first decide how often the Extractor should run. Once configured, Datamorf automatically handles the timing and execution, no cron jobs or manual triggers required.

2. Source — What Data It Pulls

Next, you configure the data source from which the Extractor will pull records. Depending on the platform, this could mean:

• Selecting an object type (e.g., Contacts, Deals, Companies in HubSpot)

• Writing a SQL query (for databases)

• Pointing to a specific endpoint (for APIs)

You can also add filters to refine what gets extracted. Datamorf uses the same visual condition builder used in workflows, so you can define filters such as “created after last week” or “status = active.”

Additionally, you can limit the number of records fetched per run. Datamorf automatically handles pagination (retrieving results in batches), so even large datasets can be processed efficiently.

3. Workflow — Where the Data Goes

Finally, you decide which workflow should process the extracted data.
Each record (or batch of records) is sent directly into the workflow as if it were a trigger event. From there, your normal workflow logic applies, the data can be transformed, enriched, and loaded to any destination system.

By default, Datamorf sends records one by one (1 request per second) to ensure stable delivery, but you can:

• Increase speed or concurrency for faster processing

• Enable batch mode to send groups of records together (up to 100 items per batch)

Datamorf also attaches metadata to each record (such as extraction ID, timestamp, and index), making it easy to trace or debug data later.

How to Connect an Extractor to a Workflow

There are two ways to connect an Extractor to a workflow in Datamorf:

1. From the Extractor Builder
   After setting up your Extractor (defining the schedule and source), you’ll be asked to choose the workflow you want to connect it to. Select the target workflow, and Datamorf will automatically establish the link, ensuring the extracted data is sent directly into that workflow for processing.

2. From the Workflow Builder
   When configuring how your workflow starts, you can select Add Extractor as the trigger method. This option will take you directly to the Extractor section, where you can choose an existing Extractor or create a new one to connect. Once linked, the Extractor becomes the starting point of that workflow, feeding data automatically according to its schedule.

Both methods achieve the same result, establishing a live connection where extracted data continuously flows into your workflow for transformation and delivery.

The Extractor is Datamorf’s automated data retrieval engine. It pulls structured data from your tools on a schedule and connects it directly to a workflow, where it can be transformed, enriched, and delivered anywhere, creating a fully automated, round-trip data pipeline.