What is Data Modelling in BI

Definition

Data modelling is the process of structuring data in a way that makes it easy to analyse, understand, and use in reporting.

It defines how data is organised, how different pieces of data relate to each other, and how they should be used.

 

What Data Modelling Actually Means

Raw data is often messy and difficult to work with.

  • Tables may not be connected

  • Fields may be inconsistent

  • Relationships may be unclear

Data modelling solves this by organising data into a clear structure.

It ensures that when you analyse data, you are working with something logical and reliable rather than something confusing and fragmented.

 

Why Data Modelling is Important

Without data modelling:

  • Reports can be inconsistent

  • Metrics can be calculated incorrectly

  • Analysis becomes difficult

With good data modelling:

  • Data is easier to understand

  • Relationships between data are clear

  • Reporting becomes reliable

In practice, data modelling turns data into something that can be trusted and used confidently.

 

Data Modelling in Business Intelligence

In a Business Intelligence system:

  • Data pipelines move data

  • Data warehouses store data

  • Data modelling organises data

  • Dashboards display data

Data modelling sits between the data warehouse and the dashboard.

It is what makes dashboards work properly.

 

Tables and Relationships

At its core, data modelling is about:

  • Tables

  • Relationships between those tables

For example:

  • A table of orders

  • A table of customers

  • A table of products

These tables are linked together so that data can be analysed across them.

This allows questions like:

  • Which customers generate the most revenue?

  • Which products are most popular?

 

Facts and Dimensions

Data models are often built around two types of tables:

  • Fact tables — contain measurable data (metrics)

  • Dimension tables — contain descriptive data (categories)

For example:

  • Orders table (facts): revenue, quantity

  • Product table (dimensions): category, name

This structure makes it easier to analyse metrics across dimensions.

 

A Simple Example

An e-commerce business has:

  • Orders data

  • Customer data

  • Product data

A data model links these together so that:

  • Revenue can be analysed by product

  • Orders can be analysed by customer

  • Performance can be broken down by category

Without this structure, analysis would be slow and unreliable.

 

Common Misconceptions

“Data modelling is only for technical people”

Data modelling affects how everyone in the business understands data.

“The data warehouse already solves this”

The warehouse stores data. The model makes it usable.

“You can skip modelling and go straight to dashboards”

Without proper modelling, dashboards often become confusing and inconsistent.

 

Why Data Modelling Matters

Data modelling makes it possible to:

  • Analyse data across different sources

  • Ensure consistency in reporting

  • Build reliable dashboards

  • Understand relationships in data

Without it, Business Intelligence becomes difficult to manage and scale.

 

Summary

Data modelling is the process of:

  • Structuring data

  • Defining relationships

  • Making data easier to analyse

It is what turns stored data into something that can be explored, understood, and used effectively.

 

Related Topics