Thread by @prosomie

In love with Power BI! 😍✨

Power BI is a data analytics tool, and the fact that you can connect it with any data source, transform, manipulate and visualize your data all in one place is really great!

Important things you need to know about creating a data model in Power BI🧵

Before you understand how creating a data model works in Power BI, first you need to understand what kind of tables you need to have in your data model.

There are two types of tables that enables you to create a data model in Power BI and they are Fact Table and Dimension tables

✨ Fact tables: Fact tables contain observational or event data values which can be sales orders, product counts, transactional dates and times and quantities. Fact table can contain several repeated values.

E.g  repeating customer IDs in the Customer ID column.

✨ Dimension tables: Dimension tables contain the details about the data in a Fact table. Like details about the products, customers and employees. This kind of table is used to filter data in the fact table and contains a unique value for each row.

E.g. unique product keys...

in the product table and unique customer ID in the customer table.

When modeling your data in Power BI, you need a fact table and dimension tables.

A fact table is connected to the dimension tables via primary and foreign key columns as seen in the diagram.

This brings us to the types data model or schema design you can create within Power BI.

And they are Schema design and snowflake design.

✨ Star schema design is a type of data model design where your tables are arranged in a star shape with the Fact table being at the center.

Star schema design is the most popular schema design and the easiest way to simplify your data.

Snowflake design is a type of schema design where dimension tables can be connected to one another and not all directly to the fact table as it's shown in this visual.

Designing your data model cannot be complete without you understanding how the tables relate and connect to each other which is known as table relationships.

✨ Many-to-one or one-to-many relationships : this is a relationship where instances of a value in one column (CityID column in Customer table) are related to only one unique corresponding instance in another table (the CityID column in the City table).

One-to-one relationships: is one in which both tables contain unique value in the connecting column. It's advisable to avoid this type of relationship as it makes your data redundant.

Many-to-many relationship: is one where many values are common between two tables. This type of relationship is not recommend because lack of unique values introduces ambiguity in your data model.

These are the basic things you need to understand while creating your data model and ensuring your data model is optimized for performance and usability.

#powerbi #dataanalystics #datamodeling