Forem: Emmanuel Atemba

Schemas and Data Modelling in Power BI

Emmanuel Atemba — Sun, 01 Feb 2026 19:20:27 +0000

In this article we will discuss two major concepts in Power Bi.

Data Modelling
Schemas

Data Modeling

Data modelling is the process of identifying, organizing and defining the types of data a business collects and the relationships between them. It uses diagrams, symbols and textual definitions to visually represent how data is captured, stored and used.

A well-designed data model helps:

Understand data requirements
Ensure proper structure for reporting
Align with business goals
Maintain data integrity

Schema

A schema is the structure of your data model. It shows what tables you have, which columns exists in individual tables and how the tables relate to each other.

In Power Bi, the type of schema you use will determine:

Report correctness
Performance
DAX Complexity and
Scalability of your model

Types of Schemas in Power Bi

There are two primary schema types in Power Bi:

The Star Schema
The Snowflake Schema

The Star Schema

The Star Schema design has a:

Fact Table: which is a central table (usualy very large) which that contains all the quantitative data from observations or events and includes foreign keys referencing dimension tables.

Dimension tables: contain descriptive attributes for business entities related to the facts table. These are the things you will be modeling, and each dimension table is connected to the fact table via a primary key. It's worth noting that these tables are denormalized

A star schema is designed to simplify relationships and is built for fast querying in analytical tools.

Key Characteristics of Star Schema

One central fact table: Which contain all the transaction information and foreign keys to other tables.
Multiple dimension tables: Which contain descriptive attributes used for filtering, grouping, and slicing data.
One-to-many relationships: Where each record in a dimension table can be associated with many records in the Fact table. Example a customer placing multiple orders.

Benefits of Star Schema in Power BI

Performance Optimization: Queries in Power BI are faster because the star schema minimizes the number of joins and simplifies relationships.
Ease of Use: Dimension tables enable intuitive filtering and grouping, while fact tables support summarization.
Scalability: The structure is flexible and can accommodate additional dimensions or facts without significant redesign.

The Snowflake Schema

The Snowflake Schema is a type of data model which is like the Star Schema in terms of having a central fact table that everything passes through, but the main difference is that the Snowflake Schema has normalized dimension tables as opposed to denormalized dimension tables for the Star Schema.

So, the Star Schema typically goes down two levels of data hierarchy [Fact Table → Dimension Table] while the Snowflake Schema goes to a third level [Fact Table → Dimension Table → Smaller Dimension Table]. These tables are joined using one-to-many relationship just like in Star Schema.

This creates a hierarchical structure resembling a snowflake, hence the name. It is primarily used to handle complex data structures while maintaining data integrity.

Key Features of Snowflake Schema

Normalization: Dimension tables are normalized into multiple related tables to reduce redundancy and improve consistency.

For example, a Customer table might link to a City table, which in turn links to a Country table.

Hierarchical Structure: The schema is built around a central fact table surrounded by normalized dimension tables, creating a multi-level hierarchy.

Joins: Queries often require more joins due to the normalized structure, which can impact performance on large datasets.

Scalability: While the schema is scalable for large datasets, its complexity can make it harder to manage and query efficiently.

Advantages of Snowflake Schema

Data Integrity: Normalization ensures that data is consistent and reduces redundancy.

Efficient Storage: It uses less disk space compared to denormalized schemas like the Star Schema.

Detailed Analysis: The hierarchical structure allows for detailed drill-down analysis.

Disadvantages of Snowflake Schema

Complexity: The normalized structure makes the schema harder to understand and maintain.

Performance Overhead: More joins are required, which can slow down query performance.

Limited Use Cases: Snowflaking is generally avoided unless necessary, as the space savings are often insignificant compared to the overall data warehouse size.

Here's is 1-on-1 comparison of the two schema types

Star Schema vs Snowflake Schema (Key Factors Only)

Factor	Star Schema	Snowflake Schema
Performance	High – fewer joins, faster queries	Lower – more joins slow queries
Scalability	Scales well as data & users grow	Scales poorly for analytics workloads
Storage efficiency	Moderate – some duplication	High – normalized, less duplication
Memory usage (VertiPaq)	Efficient compression	Less efficient compression
DAX complexity	Low – simple, readable DAX	High – complex relationships
Maintenance	Low – easy to extend & fix	High – fragile as model grows
Filter behavior	Predictable	More complex & error-prone
Enterprise BI suitability	Excellent (default choice)	Limited / niche
Power BI recommendation	Strongly recommended	Avoid unless unavoidable

Importance of Good Data Modelling in Power BI

Good data modelling is the foundation of effective Power BI reporting. It determines whether insights are accurate, fast, scalable, and trusted across the organization.

1. Ensures accurate and consistent results

A well-designed data model:
Prevents double counting
Ensures filters behave correctly
Produces consistent totals across all visuals

Without good modelling, reports can show conflicting numbers, which quickly erodes trust.

2. Improves performance

Power BI’s engine is optimized for:

Star schemas
Fewer relationships
Low-cardinality dimensions
Good modelling results in:
Faster report loading
Snappier visuals

Better performance under high user load

3. Simplifies DAX calculations

With a clean model:

Measures are shorter and easier to read
Logic is reusable
There is less need for complex workarounds

Poor modelling forces complex DAX, increasing errors and maintenance effort.

4. Enables scalability

A good data model scales when:

Data volume grows
New dimensions or KPIs are added
More users access reports

This allows the solution to grow without frequent redesigns.

5. Supports enterprise governance and security

Proper modelling:

Enables Row-Level Security (RLS)
Centralizes business logic
Supports shared and certified datasets

This ensures controlled, consistent, and secure analytics.

Improves maintainability and collaboration

Well-structured models:

Are easier to understand
Follow predictable patterns
Can be maintained by multiple analysts

This reduces dependency on individuals and supports team collaboration.

7. Encourages reuse and standardization

Good modelling allows:

One dataset to support many reports
Standard KPIs across departments
A single source of truth

This reduces duplication and inconsistency.

Final takeaway
Good data modelling is what transforms Power BI from a simple reporting tool into a reliable, enterprise-grade analytics platform

Learn Excel in a week!

Emmanuel Atemba — Sun, 25 Jan 2026 15:26:58 +0000

Everyone in the tech field or at university has at least heard of Microsoft Excel. However, the extent to which Excel is used—and how powerful it becomes—largely depends on how well you understand it. Some people use Excel simply to display data, such as sales figures or transaction records, while others go further by uncovering insights through charts, graphs, and advanced summaries.
In this article, I aim to highlight what Excel can truly help you achieve. Excel is more than a tool for displaying data; it is a powerful platform for data storytelling and decision-making when used effectively.

Table of Content

Getting started with excel software
Formulas in Excel
Working with Tables in excel
Working with Pivot Tables in excel
Charts and Graphs
Beyond the norm: Building excel dashboards

1. Getting Started with Excel

Before diving into formulas and dashboards, it’s important to understand the basic building blocks of Excel.

Key Parts of Excel

Workbook – The Excel file itself
Worksheet – Individual tabs (e.g., Staging, Analysis, Dashboard)
Cells – The intersection of rows and columns (e.g., A1, B2)
Ranges – A selection of multiple cells (e.g., A1:D20)
Formula Bar – Where formulas are written and edited
Ribbon – The toolbar containing commands (Home, Insert, Data, etc.)

In most real-world scenarios, data first lands in Excel in a raw, unstructured format—just like the Staging sheet. The goal is to transform this raw data into insights.

2. Formulas in Excel (With Real Sales Data Examples)

Let’s assume the Staging sheet contains raw sales data with columns similar to:

OrderID, OrderDate, RequiredDate, Region, 
Country, City,CustomerSegment   Channel, 
Salesperson, ProductCategory, SKU, UnitCost, 
UnitPrice   DiscountPct, Quantity

We’ll convert this data into an Excel Table named SalesData.
This allows us to write clean, readable formulas using column names.

Simple but Powerful Excel Formulas

1. Total Revenue Generated

Business Question: How much revenue have we generated in total?

> =SUM(SalesData[Revenue])

2. Average Order Value

Business Question: On average, how much does a customer spend per order?

=AVERAGE(SalesData[Revenue])

3. Total Units Sold

Business Question: How many items have we sold overall?

=SUM(SalesData[Quantity])

4. Number of Orders

Business Question: How many sales transactions were completed?

=COUNTA(SalesData[OrderID])

5. Earliest and Latest Sale Date

Business Question: What is the time range of our sales data?

=MIN(SalesData[OrderDate])

=MAX(SalesData[OrderDate])

Intermediate & Advanced Excel Formulas for Sales Analysis

These formulas answer real analytical questions businesses care about.

6. Revenue by Category

Business Question: How much revenue does each product category generate?

=SUMIFS(
  SalesData[Revenue],
  SalesData[Category],
  A2
)

Where A2 contains a category name (e.g., Electronics).

7. Revenue by Region

Business Question: Which region is performing best?

=SUMIFS(
  SalesData[Revenue],
  SalesData[Region],
  A2
)

8. Number of Orders per Customer

Business Question: Who are our most frequent customers?

=COUNTIFS(
  SalesData[CustomerName],
  A2
)

9. Monthly Sales Trend

Business Question: How do sales change month over month?

First, extract the month:

=TEXT([@OrderDate],"mmm-yyyy")

Then calculate monthly revenue:

=SUMIFS( SalesData[Revenue], SalesData[Month], A2 )

10. Calculate Revenue (If Not Provided)

Business Question: How do we derive revenue from raw transaction data?

=[@Quantity] * [@UnitPrice]

📌 Best practice: Always calculate revenue rather than manually entering it.

11. Identify High-Value Orders

Business Question: Which orders exceed a certain revenue threshold?

=IF([@Revenue] > 100000, "High Value", "Standard")

12. Handle Missing or Incorrect Data

Business Question: How do we avoid formula errors in reports?

=IFERROR(
  [@Quantity] * [@UnitPrice],
  0
)

13. Lookup Product Price or Category

Business Question: How do we enrich sales data from a master product list?

=XLOOKUP(
  [@Product],
  Products[ProductName],
  Products[UnitPrice]
)

3. Tables + Formulas = Faster Analysis

Once your formulas are written inside a Table:

They automatically copy to new rows

They update when new sales data is added

They integrate seamlessly with Pivot Tables and charts

-> Calculated columns inside Excel Table

-> Automatically filled to all data rows

4. Pivot Tables: When Formulas Are Not Enough

While formulas answer specific questions, Pivot Tables help answer exploratory questions, such as:

Which category drives the most revenue?

How do sales vary by region and month?

Who are the top 10 customers?

Pivot Tables often replace dozens of SUMIFSformulas with a drag-and-drop interface.

Pivot Table summarizing revenue by category and region

5. Charts: Turning Answers into Stories

Once formulas and Pivot Tables provide answers, charts help communicate them:

Line chart → Monthly sales trend

Column chart → Revenue by category

Bar chart → Top customers

6. Dashboards: From Analyst to Decision-Maker

A dashboard built on:

Excel Tables

Pivot Tables

Slicers

Charts

Allows leadership to answer questions like:

Are sales improving or declining?

Which regions need attention?

What products drive profitability?

Key Takeaway

Excel formulas are not just calculations—they are questions translated into logic.

If you can clearly ask:

What happened?

Why did it happen?

What should we do next?

Then Excel becomes one of the most powerful data analysis tools you already own.