Online analytical processing (OLAP) is a method of querying and analysing large quantities of data from multiple perspectives. It commonly operates on historical data stored in a data warehouse, allowing users to identify trends, comparisons and patterns quickly.
OLAP structures data conceptually as a multidimensional data cube. A cube can have more than three dimensions; the term describes data organised by several categories rather than a literal three-dimensional shape.
For example, a retailer could analyse sales using:
- Dimensions: time, product and region
- Measures: sales revenue, quantity sold and profit
The system may calculate total sales for each product, month and region in advance. These aggregated values make analytical queries faster than repeatedly calculating results from individual transaction records.
A typical OLAP process works as follows:
- Data is extracted from operational databases and loaded into a data warehouse.
- The data is organised by dimensions and measures.
- Users apply OLAP operations to explore different summaries and levels of detail.
| OLAP operation | Purpose and example |
|---|---|
| Slice | Selects one value from a dimension, such as sales during 2026. |
| Dice | Selects a subset across several dimensions, such as two products in three regions. |
| Drill-down | Moves from a summary to more detailed data, such as year to month. |
| Roll-up | Aggregates detailed data into a summary, such as city sales into national sales. |
| Pivot | Reorients dimensions to present the data from another perspective. |
A common misconception is that OLAP processes everyday transactions. That is online transaction processing (OLTP). OLTP records frequent operations such as purchases, while OLAP supports complex analysis and decision-making.