Knowledge base

ETL, reverse ETL, and iPaaS: the plumbing

Data is useless where it cannot be reached. These four acronyms are the pipes that move customer data to where the work happens. Direction is the whole story.

6 min read · Updated June 28, 2026

Filling the warehouse: ETL and ELT

ETL (extract, transform, load) and ELT (extract, load, transform) both move data into the warehouse from your sources. They differ only on when the cleaning happens: ETL transforms the data in transit before loading it, while ELT loads the raw data first and transforms it inside the warehouse using the warehouse compute. ELT is the modern cloud default because warehouses are now cheap and powerful enough to do the work.

Either way the direction is the same: sources to warehouse. This is how the warehouse gets filled with everything the company knows.

Activating it: reverse ETL

Reverse ETL runs the other way: warehouse to SaaS tools. It takes the clean, modeled customer data sitting in the warehouse and syncs it back out into the operational tools that marketing, sales, and support actually use, including the engagement platform. The crisp line from the vendors: ELT writes to the warehouse; reverse ETL reads from it.

Why this one matters most

Reverse ETL is the mechanism that makes a composable CDP possible. It is how unified warehouse data gets activated into a tool like Iterable so it can be used in a message. Without it, the warehouse is a library no one can check books out of.

Wiring apps: iPaaS

iPaaS (integration platform as a service) is the general-purpose connective tissue between applications. Gartner describes it as a cloud suite for building and running integration flows that connect applications, data, and processes. In plain terms, it wires many apps together with workflows and real-time triggers (Workato, MuleSoft, Boomi; Zapier is the lightweight tier).

The difference from ETL is intent. iPaaS moves transactions and triggers between applications, usually in real time and often one record at a time, to drive a business workflow. ETL and reverse ETL move datasets to and from the warehouse, usually in batch, for analytics and activation. The lines blur as iPaaS tools add data features, but the primary job is the tell.

Frequently asked questions

Is reverse ETL just ETL backwards?

Directionally yes. ETL and ELT move data into the warehouse; reverse ETL moves modeled data out of the warehouse into the SaaS tools that act on it. The reverse is the direction.

When would I use iPaaS instead of reverse ETL?

Use iPaaS for real-time, app-to-app workflows (when a deal closes in the CRM, create an invoice in the billing tool). Use reverse ETL to sync modeled audiences and traits from the warehouse into marketing and sales tools.

How does this reach Iterable?

A reverse-ETL tool such as Hightouch or Census, or a CDP, pushes the unified customer profile into Iterable, where it becomes available for segmentation, journeys, and personalization.

Keep reading

CDP vs data warehouse vs ESP: who does what Three tools get confused constantly because they all "hold customer data." The difference is what each is for: storing everything, unifying the customer, or sending the message. What is a Customer Data Platform (CDP)? A CDP is the tool that turns a pile of scattered customer data into one trustworthy profile per person, and makes that profile available to every other tool. Here is what that means and where the line sits. What is MarTech? A map of the customer-engagement stack Marketing technology is just the set of software a company uses to find, understand, and talk to its customers. The jargon hides a simple shape: collect data, unify it, decide what to say, and deliver it. This is the map.

Sources

Part of the MarTech and customer engagement field notes on this site. Written by Antoni K Pestka.