Learn AI · Level 2 · How AI works · step 6 of 19 · word 2 of 4
What is ETL?
ETL stands for extract, transform, load, the classic three-step recipe for moving data. Copy it from its sources, clean and reshape it, then save it where people and AI can use it.
In plain English
Extract: copy data out of the places it lives, such as a shop’s till system, its website and its app. Transform: clean it and make it match, so “UAE”, “U.A.E.” and “United Arab Emirates” become one value. Load: save the result in a data warehouse or another store.
An example
A retailer extracts each day’s sales from 200 stores, transforms every price into one currency and loads the totals into its warehouse. A demand-forecasting model reads them the next morning.
Watch out
Many teams now use ELT: they load the raw data first and transform it later, inside the store. That keeps the original data, so mistakes in the cleaning step can be fixed and re-run. Either way, a data pipeline is only as good as its transform rules.