Data Modeling in Power BI

Sara Kutty
St Joseph’s University
What Is Data Modeling?
• Data modelling is the process of analysing and defining all the
different data types your business collects and produces, as well as
the relationships between those bits of data.
• By using text, symbols, and diagrams, data modelling concepts create
visual representations of data as it’s captured, stored, and used at
your business.
• As your business determines how data is used and when, the data
modelling process becomes an exercise in understanding and
clarifying your data requirements.
What Is Data Modelling?
• Data Modeling is one of the features used to connect multiple data
sources in BI tool using a relationship.
• A relationship defines how data sources are connected with each
other and you can create interesting data visualizations on multiple
data sources.
• With the modeling feature, you can build custom calculations on the
existing tables and these columns can be directly presented into
Power BI visualizations.
• This allows businesses to define new metrics and to perform custom
calculations for those metrics.
 The benefits of data modelling
• Creating a structure for collaboration between your IT and business
teams.
• Revealing opportunities for improving business processes by defining
data needs and uses.
• Saving time and money on IT and process investments through
appropriate planning.
• Reducing errors (and error-prone redundant data entry) while
improving data integrity.
• Increasing the speed and performance of data retrieval and analytics
by planning for capacity and growth.
• Setting and tracking target key performance indicators tailored to
your business objectives.
Types of Data modeling concepts

• Conceptual data modelling

• Logical data modelling
• Physical data modelling
 Types of Data modelling concepts
• Conceptual data modelling
• defines the overall structure of your business and data.
• the foundation of every data model that’s created
• It’s a hierarchical view of entities and their relationships,
and it’s usually created to give stakeholders a broad
overview of the database
• Used for organising business concepts,
• your conceptual data model is defined by your business
stakeholders and data engineers or architects.
• They help you understand which entities exist in your
business and how they relate to each other.
• Conceptual models don’t include the details regarding the
specific attributes attached to an entity.
• For instance, you may have customer, employee, and product
data and each entities, has relationships with other entities.
Both the entities and the entity relationships are defined within
your conceptual data model.
 Types of Data modelling concepts
• Logical data modelling
• builds upon the conceptual data model with specific
attributes of data within each entity and the relationships
between those attributes.
• focuses on how data is stored in an organization’s systems.
• The logical model describes how data moves between its
source (for example, a person or another system) and its
destination (for example, a database).
• It uses entities, attributes, relationships, cardinality, and
constraints to describe the entity set for each table in a
relational database.
• For instance, Customer A buys Product B from Sales
Associate C.
• This is your technical model of the rules and data structures as
defined by data engineers, architects, and business analysts, helping
drive decisions about what physical model your data and business
require.
 Types of Data modelling concepts
• Physical data modelling
• A physical data model is your specific
implementation of the logical data model
created by database administrators and
developers.
• It is used to create tables, indexes, and views,
which are implemented through the use
of Structured Query Language (SQL)
statements.
• It is developed for a specific database tool and
data storage technology, and with data
connectors to serve the data throughout your
business systems to users as needed.
• This is the “thing” the other models have
been leading to—the actual implementation
of your data estate.
What Is Data Modelling?

• Process of creating a data model.

• Data Models
• abstract data that defines the data structure, properties and relation.
• To create a Data Model, multiple entities or values for a given scenario
must relate to each other.
• Types of data models
• Hierarchical Model,
• Relational Model,
• Network Model,
• Entity-Relationship Model etc
• For Power BI, we will be more focused on the Entity-Relationship Model.
Cardinality
Cardinality
Cardinality
Steps to Create Relationships
• To create and manage relationships in Power BI Desktop:
• On the Home tab, select Manage Relationships > New.
• In the Create relationship dialog box, in the first table drop-down list, select a
table. Select the column you want to use in the relationship.
• In the second table drop-down list, select the other table you want in the
relationship. Select the other column you want to use, and then choose OK.
 Handling Data from Multiple Sources
Handling data from multiple sources with different granularity levels involves the
following steps:
• Data Integration:
• Use Power Query to connect and extract data from multiple sources.
• Apply data transformations, clean, and prepare the data for analysis.
• Data Modeling:
• Create relationships between tables to link data from different sources.
• Use common dimensions to establish relationships between tables with different granularity
levels.
• Granularity Adjustment:
• Create new calculated columns or measures to adjust granularity levels.
• Use DAX (Data Analysis Expressions) to aggregate or disaggregate data as required.
• Data Validation:
• Compare aggregated data with source data to ensure accuracy.
• Address any discrepancies to maintain data integrity.
 Optimizing Data Model Relationships
• Minimize the Number of Relationships:
• Reduce the number of relationships by removing unnecessary ones and
consolidating tables when possible.
• Use Single-Direction Relationships:
• Avoid bi-directional relationships, as they can cause ambiguity and performance
issues.
• Use Star Schema:
• Organize data into a star schema with a central fact table and related dimension
tables to simplify relationships and improve performance.
• Optimize Data Types:
• Use appropriate data types for columns, and avoid using high cardinality columns as
keys.
• Minimize Calculated Columns:
• Replace calculated columns with measures where possible to reduce storage and
processing overhead.
Questions
• What Is Data Modeling?
• Write the different ways to create relationship.
• How do you identify a Fact table & a Dimension table while creating data modeling
• How do you identify an active relationship and an inactive relationship. How many active and inactive
relationships can be there between two tables.

• Types of Data modeling concepts

• What all points should be considered while creating a data model.
Or
What are the key considerations for developing a robust data modeling strategy within Power BI?
• Differentiate between Active and Inactive relationship in Power BI
• Differentiate between Fact and Dimension Table.
• How to create and manage relationships in Power BI Desktop?
• How do you optimize data model relationships for better performance?

• How would you handle data from multiple sources with different granularity levels?
https://powerbi.microsoft.com/en-in/what-is-data-modeling/

https://learn.microsoft.com/en-us/power-bi/guidance/relationships-
active-inactive
https://youtu.be/JQ9wX1d7dGE?si=gHFo_P-5GyE1R4kI

Data set download - https://drive.google.com/drive/folder