💡

Beginner Explain.

If Excel is a notebook where you calculate and organise numbers, Power BI is the presentation screen that turns those numbers into a living, interactive picture. Imagine showing your manager a wall of numbers versus showing them a colourful dashboard where clicking 'Lagos' instantly updates every chart to show only Lagos data. That second experience is what Power BI is built for.

💡

Beginner Explain.

Installing Power BI Desktop from the Microsoft Store is just like installing any app on your phone from an app store — you search for it, tap install, and it downloads and updates itself automatically from then on. No technical setup, no admin password needed, and Windows keeps it current in the background without you having to think about it.

💡

Beginner Explain.

Think of Power BI like a TV production. Power BI Desktop is the editing studio where you put the show together — cutting footage, adding graphics. Power BI Service is the broadcast network that sends the finished show out to viewers. Power BI Mobile is the TV in someone's pocket where they watch the show wherever they are. Each piece has a different job, but together they deliver one experience.

💡

Beginner Explain.

Think of this workflow like cooking a meal. Get Data is going shopping for ingredients. Transform is washing, peeling, and chopping them so they're ready to use. Model is the recipe — deciding how the ingredients combine and in what proportion. Visualize is plating the finished dish so it looks appetising. And just like a meal, if your ingredients change next time, you go through the same steps again — that's the Refresh loop.

💡

Beginner Explain.

Think of Power BI as a TV and a data connection as the cable plugging it into a signal source. The TV (Power BI) is just a screen until you plug in a cable box, an antenna, or a streaming device. Get Data is how you choose what to 'plug in' — your Excel file, a database, or a website — and once connected, your data starts flowing into Power BI just like a channel starts playing on the screen.

💡

Beginner Explain.

Imagine borrowing a library book. Import mode is like photocopying the whole book and taking the copy home — you can read it anytime, even without internet, and flipping pages is instant. But if the library updates the book, your copy doesn't change until you photocopy it again (refresh). DirectQuery is like reading the book through a live video call with the librarian holding it — you always see the current page exactly as it is right now, but every single page turn depends on that live connection, so it can feel slower.

💡

Beginner Explain.

Imagine receiving a box of mixed vegetables straight from a farm — some still have soil on them, some are oddly shaped, and they're all jumbled together. Before you can cook with them, you wash, peel, and sort them into separate bowls. Data structuring is exactly this, but for data: you take a messy export from Excel or a database and wash it, split it, sort it, and reshape it so it's ready to be 'cooked' into a report.

Area

Location

What It Does

Queries pane

Far left

Lists every table you've connected to — click one to edit it

Data preview grid

Centre

Shows your data exactly as it will look once loaded — edit it live here

Applied Steps

Far right

Every click you make is recorded here, in order, as an undoable, re-playable step

⚡

Why This Matters

The Applied Steps list is the single most important concept in Power Query. Because every action is recorded as a named step, your entire cleaning process can be replayed automatically the next time you refresh — even on a completely new batch of data. You clean once; Power BI repeats your work forever.

💡

Beginner Explain.

Imagine a kitchen where every utensil is just thrown into one giant drawer. Finding a specific spoon takes forever, and nobody can tell what's clean or dirty. Data management is organising that drawer — putting forks together, labelling containers, throwing out duplicates. A well-managed Power BI file is one where any colleague can open it and immediately understand what each query and table does, just from its name and organisation.

💡

Beginner Explain.

Building a report is like planting a garden — that's the exciting part. But a garden needs ongoing watering, weeding, and checking for pests, or it withers. Data activities are the watering and weeding of Power BI: refreshing the data regularly, checking that nothing has broken, and fixing small issues before they become big ones.

💡

Beginner Explain.

Imagine measuring a person's height. You could record it to the nearest metre, the nearest centimetre, or the nearest millimetre — each is a different 'granularity' of the same measurement. In a Sales fact table, granularity might be 'one row per order' (coarse) or 'one row per product within an order' (fine). The finer the granularity, the more flexible your analysis — but also the more rows you store.

Granularity Level

Example Row

What You Can (and Can't) Analyse

One row per order

Order #1001, Total = ₦45,000

Can analyse order totals — CANNOT see which products were in the order

One row per order line

Order #1001, Solar Panel, Qty 2, ₦40,000

Can analyse by product, quantity, AND order — most flexible, more rows

One row per day per store

01/06/2026, Store A, Total = ₦230,000

Good for trend reports — CANNOT see individual transactions

⚠️

Golden Rule

Always choose the FINEST granularity your source data can reasonably provide. You can always summarise fine-grained data into a coarser view using DAX or visuals — but you can NEVER break coarse data back into detail that was never captured. Storing 'one row per order line' costs more rows, but keeps every future analysis question answerable.

Cardinality Type

Meaning

Example

One-to-many (1:*)

One row on one side can match many rows on the other — the most common type

One Product can appear in many Sales rows

One-to-one (1:1)

Exactly one row on each side matches exactly one row on the other

One Employee has exactly one Employee ID record

Many-to-many (*:*)

Multiple rows on both sides can match multiple rows on the other — use sparingly

Students can belong to many Clubs, and Clubs can have many Students

💡

Beginner Explain.

In Excel, a formula like =A2*B2 always means 'multiply THIS specific cell by THAT specific cell' — it is anchored to a fixed cell address. In DAX, a formula like SUM(Sales[Amount]) means 'add up the Amount column' — but WHICH rows of Amount get added depends entirely on what filters are currently active. The same DAX formula can return a different number for every single cell in a report, because each cell applies its own filters.

💡

Beginner Explain.

Learning DAX is a bit like learning to cook. Understanding row context and filter context (from the last lecture) is like understanding how heat changes food — the underlying physics. DAX Fundamentals is more like learning your knife skills and the names of your spices: SUM, AVERAGE, IF, FILTER. You won't cook a great meal from physics alone, and you won't write useful DAX from concepts alone — you need both the theory and the toolkit.

💡

Beginner Explain.

Imagine a school asking 'How is this term's attendance compared to the SAME term last year?' Doing that by hand means carefully finding the matching date range from last year and adding it all up. Time intelligence functions are pre-built shortcuts for exactly these calendar-based questions — instead of manually working out which dates to compare, you just tell DAX the pattern ('same period last year') and it handles the date arithmetic for you.

💡

Beginner Explain.

Think of a slow Power BI report like a traffic jam. Something is causing congestion, but you cannot fix it by guessing — you need to know WHICH road is blocked before deciding to build a bypass. The Performance Analyzer is Power BI's traffic camera: it shows you exactly which visual, which query, and which step is taking the longest, so your effort goes exactly where it's needed instead of randomly tweaking things.

💡

Beginner Explain.

Imagine handing someone a newspaper front page versus a wall of plain typed text with the same news. The newspaper uses big headlines, photos, and columns to instantly tell you what matters most. A Power BI report page works the same way — the most important number should be the biggest and the first thing your eye lands on, just like a headline.

Design Principle

Practical Guidance

Most important number, top-left

Place your single headline KPI (Total Sales, Pass Rate) where the eye lands first

Biggest visual = most important

Size should reflect importance — don't give a minor detail table the same space as your main trend chart

Group related visuals together

Keep all regional charts near each other; keep all time-based charts near each other

Leave breathing room

Don't fill every pixel — empty space helps the eye separate one visual from the next

Limit to one main message per page

If a page tries to answer five unrelated questions, split it into multiple pages instead

💡

Beginner Explain.

Each visual type is a different tool in a toolbox, and like real tools, using the wrong one for the job causes problems. A pie chart is a screwdriver — perfect for a few large, simple shares of a whole, but useless (and frustrating) if you try to use it on twenty thin categories. Matching the visual to the question is half of good report design.

💡

Beginner Explain.

Building the visual (choosing the chart type and dragging in fields) is like building a house's structure — walls, roof, rooms. Formatting is the interior decorating afterward — paint colours, light fixtures, furniture placement. The house works either way, but decorating is what makes it feel finished and pleasant to be in.

💡

Beginner Explain.

Building a report in Desktop is like writing a letter on your own computer — only you can read it until you do something with it. Publishing is like putting that letter into the post: it travels from your machine to a shared online location where the people you choose can open and read the exact same content, without needing your computer or your copy of Power BI Desktop at all.

🎯

Your Mission

Using any dataset you've already built in this course (the student register, sales data, or similar), design and publish a 2-page interactive Power BI report that tells a clear story — not just a collection of charts.