What is the difference between ETL and ELT?
knowing where transformation runs in a pipeline.
ETL transforms before loading, on a separate engine; ELT loads raw then transforms inside a scalable warehouse. Choose ELT with cloud warehouses and large raw or schema-on-read data.
What's really being asked
This evaluates whether you understand not just the acronyms but the architectural shift behind them: where transformation compute runs and why cloud warehouses changed the default.
The full answer
ETL extracts data from sources, transforms it on a dedicated processing engine or staging server into the target's clean schema, then loads the finished result into the warehouse. This was standard when warehouse storage and compute were expensive and tightly coupled, so you cleaned data before paying to store it. ELT extracts data and loads it raw into the warehouse first, then transforms it in place using the warehouse's own scalable compute, often expressed as SQL. ELT became practical with cloud warehouses that separate storage and compute and scale elastically. ELT keeps the raw data available for reprocessing under new requirements, supports schema-on-read for semi-structured data, and shifts transformation work to a powerful engine you already pay for.
The mistakes people make
Saying the only difference is the order of letters, missing that the location of transformation compute and the availability of raw data are the real distinctions. Claiming ELT is always better; ETL still suits sensitive data needing masking before landing, or strict pre-load validation. Ignoring data governance, since loading raw can mean loading sensitive fields untransformed.
What usually comes next
How does separating storage and compute enable ELT? When is ETL preferable for compliance or PII masking? What role do tools like dbt play in ELT? How does schema-on-read versus schema-on-write relate?
A concrete example
A team landing terabytes of raw JSON event logs into a cloud warehouse loads them as-is, then runs SQL transformations to build clean models on demand. ELT wins here because the warehouse's elastic compute handles the heavy transforms, and keeping raw logs lets analysts rebuild models when definitions change, without re-extracting from source.
Interview question
What is the defining architectural difference that makes ELT attractive with modern cloud data warehouses?
- a.ELT loads raw data first and runs transformations using the warehouse's own scalable compute, keeping raw data for reprocessingCorrect
- b.ELT simply reverses the spelling of ETL with no real change
- c.ELT always transforms data before it ever touches the warehouse
- d.ELT removes the need to extract data from sources
Why? this is the answer
ELT loads raw data then transforms in place using the warehouse's elastic compute, preserving raw data for reprocessing. The other options either trivialize the difference, drop extraction, or describe ETL's pre-load transform.
Just read this? Test yourself on what you have been reading.
Read the original → en.wikipedia.org
- #data-engineering
- #etl
- #elt
- #data-warehouse
- #pipelines
Put your scrolling time to good use
Learn one idea, try a quiz and save useful cards for revision. Tezvyn makes it easy to learn and stay current in your tech field, a few minutes at a time.
The iPhone app is on the way
We are building it. Until it lands, nothing here is held back from you: every interview card, your saved cards, streaks and the job board all work in Safari, plus hundreds of free practice quizzes of thirty questions each. Sign in and it all carries over to the app the day it arrives.
Want it as an icon? Tap Share at the bottom of Safari, then Add to Home Screen. It opens full screen and the cards you have read stay available offline.
We are hiring for this. Open roles that interview on data-engineering — each one lists the topics its interview covers.
See open roles