A star schema is a way of arranging a Power BI data model into two kinds of table: fact tables, which hold the events being measured, and dimension tables, which hold the descriptive attributes used to filter and group those events. It is the model design Microsoft recommends for Power BI.
What a star schema is
A star schema is a pattern for organising the tables behind a report. Rather than one giant table, the data is split into a central table of events and a ring of smaller tables that describe those events. Microsoft's star schema guidance for Power BI calls it "the recommended modeling approach".
The name comes from the shape. Draw the central table in the middle and the descriptive tables around it, joined by lines, and the diagram looks like a star. Those two table types have proper names: fact tables in the centre and dimension tables around the outside.
A star schema is not a Power BI feature you switch on. It's a design choice you make when you build the semantic model, the Power BI data model, deciding which tables hold what. Get that shape right and the rest of Power BI works with you rather than against you.
Fact tables: the events you measure
A fact table holds the things that happened: the events or observations you want to analyse. In a sales report, each row of the fact table is one transaction, with figures such as quantity sold and revenue alongside it.
Fact tables are usually the largest tables in the model, because a busy business generates a lot of events. They are mostly numbers: the measurable values you'll add up, plus a few key columns that link each row to the descriptive tables. There's little descriptive text in a fact table itself.
Dimension tables: the detail you slice by
A dimension table holds the descriptive detail: the "by what" you want to break the numbers down by. Common dimensions are Date, Product, Customer and Region. Each answers a question such as "sales by month" or "sales by product category".
Dimension tables are usually smaller than the fact table and full of descriptive columns: a Product table carries the product name, its category, its colour and so on. When you filter a report or drop a field onto a chart axis, you're nearly always using a dimension. The fact table supplies the numbers; the dimension tables give you the labels and the filters.
Why Power BI is built for the star schema
Power BI doesn't just tolerate the star schema; it's tuned for it. Three things all perform best when the model is shaped this way: the storage engine that compresses and holds the data, the relationships that carry a filter from one table to another and the way calculations are worked out.
Calculations are a clear example. Microsoft's DAX overview describes DAX, the formula language Power BI uses, and DAX measures evaluate fastest and most predictably over a clean star schema. When a filter on a dimension flows cleanly to the fact table, a measure has a simple, fast path to follow.
The practical payoff is reports that stay quick, files that stay small and measures that behave the way you expect. A model that fights the star schema tends to be slow and unpredictable instead.
The star schema versus the alternatives
The most common alternative is a single flat table: every column in one wide sheet, much like a large Excel export. For a tiny dataset that's fine. At any real scale it slows down, repeats the same descriptive text on thousands of rows and makes calculations harder to write and filtering harder to predict.
A snowflake schema is the other contrast. It still has fact and dimension tables, but the dimensions are normalised further into linked sub-tables, so there are more relationships to hop through. In Power BI a star schema is generally preferred: it's simpler to build and to reason about while giving the same answers. Shaping the star is a loading job. The Power Query step is where tables are split and joined into the star shape before they reach the model.
A good model is where good Power BI begins Reading about the star schema is a useful start; designing one from your own messy data is a skill worth building properly. Our two-day, hands-on Power BI Masterclass teaches data modelling the practical way, shaping real tables into a star schema rather than learning the theory in the abstract. If you are weighing up where to learn, our Power BI training buyer's guide compares the UK options.
