Reverse ETL is an important part of the Modern Data Stack (MDS) today. Basically it enables to bring back your analytical data in a operational context to make active use of them where business happens. This is typically called Data Activation1 and is often seen in the context of marketing and sales applications.
As Reverse ETL is a typical MDS topic it is considered as part of MDS 2.0.
This is part of my “The Modern Data Stack” blog series, focussing on the Reverse ETL part.
Basically in classical data stacks like SAP, Reverse ETL is not new. Around 20 years ago SAP BW and SAP CRM worked together to enable so called closed-loop scenarios where data is brought back from the analytical context to the operational context2. Another SAP example is the retraction of controlling data from SAP BW Integrated Planning (BW-IP)3.
A general difference between ETL and Reverse ETL is that data is synced to the operational system. There are different modes:
Mirror - Destination is always on the same state as the source (Cloud Data Platform)
Upsert, Insert, Update - Insertion of a new record and update, if already exist
Further aspects are to handle change data capturing or diffs and handling errors to not sent data more than once.
Now even if in the tool market for Reverse ETL there are different approaches, in general there are some points why buying tool can make sense4:
Reverse ETL is not the same als ETL - Integration into operational apps is another thing
Tools are self-service oriented to be used by the business and make thinks like data de-duplication easier
Cost of building and maintaining an own solution can be very high
Reverse ETL can help to prevent point-to-point connections and activities trying to activate your valuable data
I can see many vendors here in the Customer Data Platforms environment but possible use cases I can see are:
Finance: optimize account payable, account receivable and cash flow
Customer: supporting marketing campains and customer contacts
Sales: make product recommendations and provide customer history information
Logistics: optimize supply chains and warehouses via vendor managed inventory and sales forecasting
In conclusion we can understand Reverse ETL rather as an SaaS - means cloud-based approach fitting well into the whole MDS idea of self-service and automation and plug-and-play thinking. A general difference to earlier approaches as shown with SAP are the self-service and automation aspects of current solutions. Nevertheless we should be aware about what it means to data governance and your data strategy.
From the timeline Reverse ETL came up since 2017/2018 as seen here with typical vendors:
Popular vendors for Reverse ETL are Octolis (2021), Rudderstack (2019), Census (2018), Hightouch (2018), Grouparoo (Airbyte, 2019) , Hevo Activate (2017), Omnata (2019), Polyatomic (2019), Seekwell (Thoughtspot, 2018)