← Lessons

quiz vs the machine

Gold1350

System Design

The ETL vs ELT Pipeline

Whether you transform data before or after loading it into the destination.

4 min read · core · beat Gold to climb

Same letters, different order

Both patterns extract data from sources and load it into a destination, but they differ on where transformation happens.

  • ETL means extract, transform, load. Data is cleaned and reshaped in a separate engine first, then only the finished tables land in the warehouse. This was standard when warehouse compute was scarce and expensive.
  • ELT means extract, load, transform. Raw data lands first, then transformations run inside the powerful warehouse using SQL. This suits cloud warehouses where compute scales on demand.

Why ELT rose

Cloud warehouses separated storage from compute and made transformation in place cheap and fast. Loading raw data first also keeps a full untransformed copy, so you can rebuild downstream tables when logic changes without re extracting from sources.

Key idea

ETL transforms before loading to save scarce destination compute, while ELT loads raw first and transforms inside an elastic warehouse, keeping a replayable raw copy.

Check yourself

Answer to earn rating on the learn ladder.

1. In ELT, where do transformations run?

2. What advantage comes from loading raw data first in ELT?

3. Why did ELT become popular with cloud warehouses?