DEV Community: David Maina

Power BI for Beginners: Understanding Relationships, Joins, Schemas, and Dashboards

David Maina — Mon, 29 Jun 2026 13:59:16 +0000

A practical guide to connecting your data and turning it into powerful visuals

Introduction

When you first open Power BI and load multiple tables, one of the first things you'll encounter is the Model view — a place where your tables sit like islands, waiting to be connected. Understanding how to connect those tables, and then turn the results into a dashboard, is one of the most important skills in data analytics.

In this article, we'll break down four foundational concepts:

Relationships — how tables talk to each other in Power BI
Joins — how rows from different tables are combined
Schemas — how your tables are organised and structured
Dashboards — how to present your insights visually

By the end, you'll have a clear mental model of how Power BI handles data behind the scenes — and how to make smarter decisions in your reports.

Part 1: What Are Relationships in Power BI?

A relationship in Power BI is a link between two tables based on a shared column — called a key. This is what allows Power BI to know that a sale in your Sales table belongs to a specific customer in your Customers table.

Why Do Relationships Matter?

Imagine you have two tables:

Customers Table

CustomerID	CustomerName	City
1	Alice	Nairobi
2	Bob	Mombasa
3	Carol	Kisumu

Sales Table

SaleID	CustomerID	Amount
101	1	5,000
102	2	3,200
103	1	7,800

The column CustomerID appears in both tables. By creating a relationship on this column, Power BI can tell you that Alice (CustomerID 1) made purchases totalling KES 12,800.

Without a relationship, Power BI treats these as completely separate tables — your visuals won't be able to combine data from both.

How to Create a Relationship in Power BI

Go to the Model view (the icon with three connected shapes on the left panel).
Drag the CustomerID column from the Customers table onto the CustomerID column in the Sales table.
Power BI creates a line between the two tables showing the relationship.

You can also go to Home → Manage Relationships → New to set one up manually.

The Model view in Power BI Desktop — tables are connected by dragging shared key columns together.

Part 2: Cardinality — The "Type" of Relationship

When you create a relationship, Power BI assigns it a cardinality, which describes how many rows in one table match rows in the other.

One-to-Many (1:*)

This is the most common relationship type. One row in Table A can match many rows in Table B.

Example: One customer can have many sales. One product can appear in many orders.

The "one" side is usually the lookup/dimension table (e.g., Customers, Products)
The "many" side is usually the fact table (e.g., Sales, Orders)

Many-to-One (*:1)

This is the same as one-to-many, just viewed from the opposite direction. Power BI treats them identically.

One-to-One (1:1)

Each row in Table A matches exactly one row in Table B — and vice versa.

Example: A table of employees and a table of employee ID cards.

Use this sparingly. If two tables are one-to-one, you could often just merge them into a single table.

Many-to-Many (:)

Multiple rows in Table A can match multiple rows in Table B.

Example: Students and Courses — one student takes many courses, and each course has many students.

Many-to-many relationships can be tricky and may cause unexpected results. As a beginner, try to avoid them until you're comfortable with the basics.

Part 3: Cross-Filter Direction

When you connect two tables, Power BI also sets a cross-filter direction — this controls how filters flow between tables when you interact with visuals.

Single Direction (→)

Filters flow from one table to another in one direction only. This is the default and recommended setting for most cases.

Example: Filtering by a product category in the Products table will filter the Sales table. But filtering in Sales won't filter Products.

Both Directions (↔)

Filters flow in both directions simultaneously.

Use "Both" with caution. It can cause ambiguous results and slow down your reports, especially in larger data models.

Part 4: What Are Joins?

A join is the operation that combines rows from two or more tables based on a related column. Joins are a concept from SQL (databases), but Power BI performs joins behind the scenes when you set up relationships or merge queries in Power Query.

Understanding joins helps you know which rows will appear in your final report.

The four main join types illustrated. Power BI supports all of these through Power Query's Merge Queries feature.

Inner Join

Returns only the rows that have a matching value in both tables.

Example: If CustomerID 4 exists in Sales but not in Customers, that sale will be excluded from the result.

Customers INNER JOIN Sales ON CustomerID
→ Only customers who have made at least one sale appear

Left Join (Left Outer Join)

Returns all rows from the left table, and the matching rows from the right table. If there's no match, the right side shows blanks/nulls.

Example: All customers appear, even those who haven't made any sales yet.

Customers LEFT JOIN Sales ON CustomerID
→ All customers appear; sales columns are blank for those with no sales

Right Join (Right Outer Join)

Returns all rows from the right table, and the matching rows from the left table.

Full Outer Join

Returns all rows from both tables, with nulls where there's no match on either side.

How to Apply Joins in Power BI (Power Query)

In Power Query Editor, select a table.
Go to Home → Merge Queries.
Choose the second table and the matching columns.
Select the Join Kind (Inner, Left Outer, Right Outer, Full Outer, etc.).
Click OK and expand the merged column to access the data.

The Merge Queries dialog in Power Query — this is where you select your join type and matching columns.

Part 5: What Are Schemas?

A schema is the overall blueprint of how your tables are organised and how they relate to each other. In Power BI, you'll typically work with one of two schema types: Star Schema or Snowflake Schema.

Star Schema ⭐

The star schema is the most recommended structure for Power BI reports. It has:

One central fact table — contains measurable, transactional data (e.g., Sales, Revenue, Orders)
Multiple dimension tables surrounding it — contain descriptive attributes (e.g., Customers, Products, Dates, Locations)

         [Date Table]
              |
[Products] — [Sales Fact Table] — [Customers]
              |
         [Stores Table]

A classic Star Schema in Power BI — one central fact table connected to several dimension tables.

Why is Star Schema great for Power BI?

Simpler relationships (mostly one-to-many)
Faster report performance
Easier to understand and maintain
DAX calculations work more predictably

Practical Example:

Table	Type	Contains
Sales	Fact	SaleID, Date, ProductID, Amount
Products	Dimension	ProductID, Name, Category
Customers	Dimension	CustomerID, Name, City
Date	Dimension	Date, Month, Quarter, Year

Snowflake Schema ❄️

A snowflake schema is an extension of the star schema where dimension tables are further broken down into sub-dimensions.

[Category] → [Products] → [Sales Fact Table] → [Customers] → [Region]

Pros:

Less data repetition (more normalised)
Smaller storage size

Cons:

More complex relationships
Slower performance in Power BI
Harder to manage

For most Power BI projects, stick to a **Star Schema. It's simpler, faster, and Power BI is optimised for it.

Part 6: The Date Table — A Must-Have

One dimension table that every Power BI model should have is a dedicated Date table. This is a table with one row per calendar day, along with columns like Month, Quarter, Year, and Week Number.

Having a proper Date table allows you to:

Use time intelligence functions in DAX (e.g., year-to-date totals, month-over-month comparisons)
Filter reports cleanly by date periods
Avoid gaps in your timeline visuals

How to Mark a Table as a Date Table

Select your Date table in the Model view.
Right-click → Mark as Date Table.
Select the column that contains unique dates.

Part 7: Power BI Dashboards

Now that your data model is set up with proper relationships and a clean schema, the final step is to visualize your insights — and this is where dashboards come in.

What Is a Power BI Dashboard?

A dashboard in Power BI is a single-page canvas that displays tiles — visual summaries of your key metrics. Think of it as your data's "control panel": you look at it and instantly know how your business or project is performing.

Dashboards are different from reports:

Feature	Report	Dashboard
Pages	Can have multiple pages	Always one page (canvas)
Interactivity	High — slicers, drilldowns, filters	Limited — click to open report
Purpose	Deep exploration and analysis	High-level overview and monitoring
Created from	Datasets directly	Pinning visuals from reports
Real-time data	Not always	Yes — supports live tiles

In simple terms: you build a **report* to analyze data, then you pin the most important visuals from that report to a dashboard for quick monitoring.*

Key Components of a Dashboard

1. KPI Cards
These show a single important number — like total revenue, number of sales, or percentage growth. They're the first thing you read on a dashboard.

2. Charts and Graphs

Bar/Column charts — compare values across categories (e.g., sales by region)
Line charts — show trends over time (e.g., monthly revenue)
Pie/Donut charts — show proportions (e.g., sales by product category)

3. Tables and Matrices
Good for showing detailed breakdowns when you need more than just a chart.

4. Slicers
Interactive filters that let viewers narrow down what they see — by date, region, product, etc.

5. Maps
Visualize geographic data — sales by city, customers by county, etc.

A sample Power BI Sales & Marketing dashboard — notice the mix of KPI tiles, charts, and trend lines.

How to Build Your First Dashboard

Step 1: Build your report first

Open Power BI Desktop.
Load your data (Excel, CSV, or database).
Set up relationships in the Model view.
Go to the Report view and start adding visuals from the Visualizations pane on the right.

Step 2: Publish to Power BI Service

Click Home → Publish.
Sign in to your Power BI account.
Choose a workspace and click Publish.

Step 3: Pin visuals to a Dashboard

Open your published report in Power BI Service (app.powerbi.com).
Hover over a visual — a pin icon (📌) will appear.
Click the pin icon → choose New Dashboard or an existing one.
Repeat for each visual you want on the dashboard.

Step 4: Arrange your tiles

In the dashboard view, drag and resize tiles to arrange them logically.
Put KPI cards at the top for quick reading.
Place detailed charts below.

Dashboard Design Tips for Beginners

Less is more — don't cram 20 visuals onto one dashboard. Pick the 5–8 most important ones.
Use a consistent colour theme — pick 2–3 colours and stick with them.
Always label your visuals — add titles so viewers know what they're looking at.
Put the most important metric top-left — that's where the eye naturally goes first.
Use slicers — let the viewer filter by date or category without leaving the dashboard.

A good dashboard answers one question at a glance: "How are we doing?"

Part 8: Common Beginner Mistakes to Avoid

1. Using Many-to-Many relationships without understanding them
Many-to-many relationships can cause double-counting and confusing results. Always try to resolve them by introducing a bridge table or restructuring your data.

2. Not having a Date table
Relying on date columns scattered across your fact table limits what you can do with time analysis. Always build or import a proper calendar/date table.

3. Relationship direction confusion
Make sure filters flow in the direction that makes sense for your analysis. Default to single-direction and only use bidirectional when absolutely necessary.

4. Duplicates in the "one" side of a relationship
If the column you're using on the "one" side has duplicate values, Power BI will throw an error or switch the relationship to many-to-many. Always ensure your dimension table key column has unique values.

5. Building the dashboard before the model
Many beginners jump straight into visuals without validating the data model. Always check your relationships in the Model view first — a broken model gives you wrong numbers in your dashboard.

6. Overloading the dashboard
A dashboard with 15 charts is just as confusing as raw data. Focus on the metrics that matter most.

Summary

Here's a quick recap of everything we've covered:

Concept	What It Means
Relationship	A link between two tables based on a shared key column
Cardinality	The type of match between rows (1:1, 1:Many, Many:Many)
Cross-filter	The direction filters travel between related tables
Inner Join	Returns only rows with matches in both tables
Left Join	Returns all rows from the left table, nulls for no matches
Full Outer Join	Returns all rows from both tables
Star Schema	One fact table surrounded by dimension tables
Snowflake Schema	Star schema with further-normalised dimension tables
Dashboard	A single-page visual summary of your key metrics
Report vs Dashboard	Reports are for deep analysis; dashboards are for monitoring

Final Thoughts

Relationships, joins, schemas, and dashboards might feel abstract at first, but they become second nature with practice. Every time you load a new dataset into Power BI, ask yourself:

What is my fact table? (the thing I'm measuring)
What are my dimension tables? (the things I'm grouping or filtering by)
How should these tables relate to each other?
Are my keys unique on the "one" side of the relationship?
What 5–8 key metrics should go on my dashboard?

Getting these foundations right means your reports will be faster, your calculations will be accurate, and your dashboards will make sense to anyone who looks at them.

Happy modelling! 🚀

*Written as part of my data analytics learning journey at LuxDevHQ Bootcamp.

From Zero to Spreadsheet: My Honest First Experience with Microsoft Excel

David Maina — Mon, 29 Jun 2026 13:09:41 +0000

A beginner's journey through cells, formulas, and a few accidental disasters

When someone first suggested I learn Excel, my reaction was something like: "Isn't that just a table? How hard can it be?"
Spoiler: I was very wrong.

The First Open
The first time I launched Excel, I was met with a grid of empty cells stretching into what felt like infinity. Columns labeled A, B, C... all the way to who-knows-where. Rows numbered 1, 2, 3... into the abyss. My first thought was: where do I even start?
I typed my name into a random cell somewhere in the middle of the sheet. Pressed Enter. It moved down. I pressed Tab. It moved right. I pressed Backspace. It deleted the cell content, not a character. I was already learning things I didn't expect to learn.

Cells, Rows, and Columns — Oh My
The basic vocabulary of Excel sounds simple until you're actually using it:

A cell is a single box, identified by its column letter and row number — so B3 means column B, row 3.
A row runs horizontally across the sheet.
A column runs vertically.

I spent a full embarrassing 10 minutes clicking around just to understand which was which. Then came the realization that everything in Excel is built on top of this grid. Every formula, every chart, every trick — it all refers back to cells.

My First Formula (and Why It Felt Like Magic)
The real "aha" moment came when I typed my first formula.
I had a small list of numbers — just five made-up values — and someone told me: "Type =SUM( and then select your numbers."
I typed =SUM(B2:B6) and hit Enter.
Excel calculated the total instantly.
I stared at the screen. I moved one of the numbers. The total updated by itself. I changed another number. It updated again. I sat there changing numbers for five minutes just because I could.
That's when I understood what Excel actually is: it's not a table. It's a live calculation engine wrapped in a grid.

The Mistakes That Taught Me the Most
No honest beginner's story skips the mistakes. Here are mine:

Typing numbers as text I once typed '42 (with an apostrophe) into a cell by accident. Excel treated it as text, not a number. My SUM formula ignored it completely. I spent 20 minutes checking my formula before someone pointed at the tiny green triangle in the corner of the cell and said: "That's your problem."
Deleting a formula column I selected what I thought was an empty column and pressed Delete. I had actually selected my formula results. Gone. The undo button (Ctrl + Z) became my best friend that day.
Dragging formulas the wrong way Excel has this incredibly useful feature where you can drag a formula across cells and it adjusts automatically. Except when you don't want it to. I had a formula referencing a fixed total, dragged it down, and watched it happily reference empty cells one by one. Learning the difference between relative (B2) and absolute ($B$2) cell references felt like unlocking a cheat code.

What Actually Clicked for Me
After a week of stumbling around, a few things started to feel natural:

AutoFill — Dragging the small green square at the corner of a cell to fill a pattern or copy a formula. An absolute game-changer for repetitive data.
Formatting — Bold headers, colored rows, and borders make a spreadsheet readable instead of just a wall of numbers. I underestimated this completely at first.
Sorting and Filtering — Selecting a column and sorting A-Z. Filtering to show only certain rows. These two features alone made me feel genuinely productive.
Basic functions — SUM, AVERAGE, COUNT, MIN, MAX. Five functions that cover a surprising number of real-world needs.

What I Wish I Knew on Day One
Looking back, here's what would have saved me a lot of confusion:

Press F2 to edit a cell instead of retyping the whole thing.
Ctrl + Z is undo, always. Use it fearlessly.
Click the column header to select the whole column — don't drag from the first cell.
Freeze your top row (View → Freeze Panes) so your headers stay visible when scrolling down.
Save often. Excel crashes happen. Ctrl + S is muscle memory now.

Excel in the Real World — Why It Actually Matters
Once I got past the basics, I started noticing Excel everywhere. It's not just a school assignment tool or something accountants use — it's woven into how businesses, teams, and individuals make decisions every single day.
Here's what genuinely surprised me about how useful it is outside of tutorials:
💼 Business & Finance
Companies use Excel to track budgets, forecast revenue, manage payroll, and analyze expenses. A small business owner doesn't need expensive software — a well-built spreadsheet can track income vs. expenses, flag when spending is too high, and project next month's numbers automatically. Those same SUM and AVERAGE formulas I learned as a beginner? That's the foundation of real financial models used in boardrooms.
📊 Data Analysis Without Code
Before I ever touched Python or SQL, Excel was already doing data analysis. You can take a messy dataset — hundreds or thousands of rows — and in minutes: sort it, filter it, find duplicates, calculate averages per category, and visualize it in a chart. For non-developers, this is incredibly powerful. For developers, it's a quick way to explore data before writing a single line of code.
🗂️ Project & Task Management
Teams use Excel to manage project timelines, assign tasks, track deadlines, and log progress. A simple spreadsheet with columns for Task, Owner, Due Date, and Status can run a small team's entire workflow. I've seen startups run their operations for months on nothing more than a shared Excel file.
🏫 Education & Research
Teachers use it to track grades and attendance. Researchers use it to log experimental data, calculate statistics, and produce charts for papers. Even survey data from Google Forms can be exported directly into Excel for analysis. It bridges the gap between collecting information and making sense of it.
🏠 Everyday Personal Use
This one hit closest to home. I built a simple personal budget tracker — income, fixed expenses, variable spending — and for the first time I could see where my money was going. No app required, no subscription, just a spreadsheet I made myself. Excel makes that possible for anyone.
The Bigger Picture
What makes Excel so effective in the real world is that it's flexible enough to fit almost any problem. It doesn't force you into a specific workflow. You design the structure, you write the logic, you decide what matters. That freedom is what makes it intimidating at first — but also what makes it so powerful once you understand it.
Learning Excel isn't just learning a tool. It's learning to think in structured, organized ways about data — and that skill transfers everywhere.

Where I'm Headed Next
I'm still a beginner. My spreadsheets are nowhere near the impressive dashboards I see online. But I've gone from being confused by a grid to actually understanding why Excel is one of the most used tools in the world.
Next on my list: learning VLOOKUP (everyone says it's essential), building a simple budget tracker, and maybe — maybe — getting into Pivot Tables.
If you're also just starting out with Excel, I hope this made you feel less alone. The confusion is normal. The mistakes are part of it. Keep clicking around.

Thanks for reading! If you found this useful, drop a reaction or leave a comment — I'd love to know what your first Excel experience was like too.

HOW EXCEL IS USED IN REAL -WORLD DATA ANALYSIS.

David Maina — Sat, 06 Jun 2026 19:58:12 +0000

At first i thought that excel was only meant for creating spread sheets and tables until i started my learning journey with LuxDevHQ and realized its much more. Excel comes around as a powerful tool used by people to uncover insights, research, organize data and most of all in businesses and financial institutions.

In most cases the real world tends to use excel as a tool to analyze data and execute better business decision making which includes customer data, sales figures, operational records, in short its basically used to collect large amounts of information every single day. Its basically the best tool to organize information, identify trends and plans future strategies.

I've realized that excel being one of the mostly used tool, its widely used in budgeting and financial reports. Businesses makes use of it by tracking income, profits, expenses and cash flows. Its the best tool to use in organizations to make informed decisions and improve the companies financial planning.

One common use of excel that not so many people talk about, is its effectiveness in marketing analysis. Most companies use it to monitor customer engagement, performance of advertising campaigns and also sales results. As a result excel brings out the comparison of different metrics, where marketers can identify the best strategies that are working and also where to improve.

During my first week of learning Excel, i was introduced to several useful features and formulas. The SUM function that really helps to calculate totals way fast, while the AVERAGE makes it simpler to understand the overall performance by finding the mean value of a data set. Something else i got to learn was sorting and filtering data, that helps users focus on important information and discover patterns easily. As simple as it may seem, they are essential for working with data effectively.

Personally i viewed data as just a collection of numbers and figures, but since joining LuxDevHQ bootcamp i have really change my perspective view on Excel and data. its safe to say that data can relate to stories, reveal different trends and support important decisions. Even though i am still a beginner, i can already see how valuable Excel is the in the field of data analytics. Completing a whole week learning about excel has really given me a very strong foundation, and i am super excited to continue building my skills and learning how to turn data into meaningful insights.

How to Prepare for a Technical Data Engineer Interview

David Maina — Thu, 28 May 2026 18:30:03 +0000

Data Engineering Landscape in Kenya

ududiudijkcjkmcxkmkxjkmskjkjdkjguyagskjhkhiusdkjsnfkhuiiugfuhuifuiiufgugfygfuyyfgygfufrufgrjfrjgyfstyudgufiugysafciudhfoi.

step by step guidance

Undestand the role clearly

jhfuywieuifuhiufwyeuhdufhiuwufhoweihihuirehfjfjdgfgfhfshhsdshuuehrjehehjhfejfjysffdfgdfdyeyugegduhfiugeiu.
common resposibilities include;

gudtyeuufffyuj
jdufgfgffgfugfgf
hfjfhgfghdhfvcvgdhhd
dgfdghdhhfddhdgffdh
hdyjfgggcgdydhvgd

for example,finding duplicate

SELECT
    customer_id,
    SUM(amount) AS total_amount
FROM trans
GROUP BY total_amount
GROUP BY TOTAL_amount DESC
LIMIT 5;

3. Learn about python for Data

jhfuyfggdhdfrcgfgfgdhdhdfdcdgdfdcdcdvddfgddhjddgdfddcdcfdfdfdfddcfdfdffdcdfdfgdgdgdfdffdcdccddggdfdccvdfdfddggdhhdhdhgdddfcd.
For example; importing pandas
importimport requests

**4. Prepare for cloud and pipline tools

msadggdgdgsffcgfdtsfdtfstfdftfdffdhhffdcsdsdscsdddfdfdfdfeejjuejdhdfddcdfsfscscdsdsfsfsddscsdfdfdfdfdfdfdffdfdcdfdffd.
Aahdjjdjjdjdjfhjfjkfkkfkfjfjfjjfjffjfjfjfjfjffjjfhfhfjfjfjjffjfjfjjfjfjjfjjfjhffgfgtffgfftfgdgdggg;

Ikehjdjjdjhdjjrjjrhdhhdhshsgdydhhdhdhfyyhfhfhffujfjfjjjdjjejhhfhhfhfhfhfhhffyyryrfyrygfgfgfgffyyyddjdjdjjdjdydyddgddfyfyfjjffyyfyjfjujfjufyytdytdtdydyduudjdjdjjjf.

Here is an flow chart of helping to understand for an interview as a data engineer.

To understand check the article below.
how to be a data analyst