ETL (Extract, Transform, Load)

ETL stands for extract, transform, load. Data is pulled from source systems, cleaned and reshaped (types fixed, duplicates removed, fields joined), then written to a destination, usually a data warehouse.

Why it matters

Source systems store data in different formats and quality levels. ETL gives analysts one consistent dataset. A related pattern, ELT, loads raw data first and transforms it inside the warehouse, which modern cloud warehouses make practical. Documenting each transformation makes numbers easier to trust and debug.

Example

A nightly job:

Extract:   SELECT * FROM orders WHERE updated_at > :last_run
Transform: convert currency to USD, drop test orders
Load:      INSERT into warehouse.fact_orders

If this job silently stops, dashboards go stale, so scheduled jobs are worth watching with cron monitoring.

Common mistakes

• Jobs that are not idempotent, so a rerun creates duplicate rows.
• No alerting on failure or on unexpectedly low row counts.
• Transformations buried in code with no documentation.

Related

• Data pipeline
• Data warehouse
• Data cleaning
• Heartbeat monitoring

This page is general information written by our editorial team and updated when documentation changes. Servers and providers differ, so test changes safely first. See our editorial policy and disclaimer.