Glossary

Slowly changing dimension

The techniques for recording how a dimension table's attributes change over time without breaking historical reporting.

Also called: SCD, SCD Type 1, SCD Type 2

A slowly changing dimension (SCD) is a pattern for updating a dimension table when a real-world attribute changes — a customer relocates, a product is recategorized, an employee changes department — without losing the ability to report history accurately. The name distinguishes these occasional attribute changes from the fact table, which grows with every new event.

The most common approaches are named by type: Type 1 overwrites the old value with the new one, losing history but keeping the table simple; Type 2 adds a new row with a new surrogate key and effective-date columns, preserving every historical version so a past report still reflects the attribute as it was at the time; Type 3 keeps limited history in extra columns, such as "previous region." Type 2 is the default choice whenever historical accuracy matters.

Getting SCD handling wrong is a frequent cause of reports that silently rewrite history — a Type 1 dimension will change last year's sales-by-region totals the moment a customer moves, because there is no way to distinguish which region applied at the time of each historical fact. Tools like dbt and dedicated data vault patterns automate SCD Type 2 logic so analysts do not have to hand-write the merge logic themselves.

Last reviewed September 22, 2026

In the index now

Related terms

Dimension table A table of descriptive attributes, such as customer or product details, that a fact table references for context. Fact table The table in a dimensional model that stores measurable, numeric events, each row linked to the dimensions describing it. Surrogate key An artificial, meaningless identifier generated by the warehouse to uniquely identify a dimension or fact row. Data vault A modeling methodology that splits data into hubs, links, and satellites to make a warehouse auditable and adaptable to change. Incremental model A transformation that updates only new or changed rows on each run instead of rebuilding the entire table. Data modeling The practice of organizing data into entities, relationships and schemas that make it accurate and efficient to query.

Related tools

Artie Managed real-time data replication that streams row-level database changes to a warehouse with sub-minute latency.

Related guides

How to choose a change data capture tool Pick CDC by how it reads the source log, who has to operate it, and whether you actually need sub-second delivery or just a fresh warehouse table.