Reverse ETL

What is reverse ETL?

Reverse ETL(Extract Transform Load) is the process of syncing data from a source of truth like a data warehouse to a system of actions like CRM, advertising platform, or other SaaS app to operationalize data.

That’s basically just a fancy way of saying reverse ETL lets you move data about your users from your warehouse and makes it available for frontline business teams to use in their favorite tools.

However, to really understand the power of reverse ETL (and why it’s not just another data pipeline), we first need to take a quick look at what traditional ETL pipelines made possible for business and data teams.

What’s in a name: ETL vs reverse ETL

The traditional extract, transform, load (ETL) data pipeline has remained largely unchanged since the 1970s: extract the data from the source, convert it to a usable format (or transformation), then load it into your data warehouse.

The advent of flexible data pipeline tools like Airbyte has also made it possible to load your data into the warehouse and then use your storage target to transform it (referred to as ELT). These ETL/ELT enabled companies to combine data from multiple sources into a single source of truth to inform business intelligence decisions.

This version of the modern data stack worked well when data sources were more limited (i.e. there was less data volume) and the data engineers who supported these stacks had ample bandwidth to process and answer questions about data. As you’ve probably experienced, that’s no longer the case and teams need more sophisticated tools to achieve the dream of operational analytics.

This reverse ETL makes operational analytics possible. Reverse ETL tools flip the Airbyte role, extracting data from the warehouse, transforming it so it plays nice with the target destination API (whether Salesforce, HubSpot, Marketo, Zendesk, or others), and loading it into the desired target app.

Modern data stack 2.0: The era of operational analytics

The growth in popularity of this new generation of data stack is emblematic of an important trend: Companies need to move data capabilities out of centralized silos and embed them within teams across business functions.

Reverse ETL equips these teams with detailed data inside the tools they're already using like Salesforce or Hubspot, empowering them to be more effective in their day-to-day work. The reverse ETL process effectively aligns your organization and applications around your source of truth. From there, business teams can build a shared, deep understanding of customers like never before.

The continuous flow of data--from raw data being pulled into apps to data being modeled to data being deployed into each app--creates a virtuous loop of operational analytics. And it’s only possible with reverse ETL.

This new genre of data tools closes the feedback loop that separated DataOps from DevOps and makes it possible for teams to deploy relatively real-time data and insights to core apps and services. - Boris Jabes, CEO at Census, The Operational Analytics Loop: From Raw Data to Models to Apps, and Back Again.

The modern data stack 2.0 generally consists of the following tools performing four key functions to close the operational analytics loop:

• Data integration: Also referred to as collection, this is
  an ETL tool like Fivetran or Snowplow that integrates your data sources into your warehouse.

• Data storage: A data warehouse that can store structured and unstructured data in one place like Google BigQuery, Snowflake, or Amazon Redshift.

• Data modeling: A modeling tool like dbt comes pre-configured with a massive library of data models to make your data usable in different situations.

• Data operationalization: A reverse ETL tool like Census will pull data out of your warehouse, validate it, and load it into applications that need it like Salesforce or Zendesk.