1. Know your audience
Welcome to this course on data visualization in Power BI. In this course, we will combine a discussion of the psychology of data visualization, along with practical tips on using Power BI visuals.
My name is Iason, and I will be guiding you through this course.
2. Dashboards versus "pixel-perfect" reports
Power BI offers two methods of showing data to users: dashboards and "pixel-perfect" reports.
Dashboards let us see and act upon critical business or operational metrics, possibly even in real-time. By contrast, reports give you more detailed information that won't fit on a dashboard.
That's because dashboards are supposed to have a "single pane of glass" feel, where everything is visible without manipulating the dashboard. However, reports are intended to be printed out, so they can span multiple pages.
Dashboards are the norm for Power BI, even when we say the term "report." We call pixel-perfect reports Paginated Reports in Power BI.
3. Your intended audience
When designing a dashboard, keep these questions in mind.
Dashboards should have specific intended audiences. A dashboard meant for a CEO will look quite different from one meant for an accountant.
Think about how they will use your dashboard. Are they sitting in front of a 4K monitor with a keyboard and mouse? Or are they looking at it on a phone as they move from one meeting to the next?
Then, what do you want them to do after looking at your dashboard? The better you understand the viewer's available actions, the better you can tailor your dashboard to give them the timely information they need to act.
Finally, are you showing the right measures in the right way?
This is an example of a process control chart, and the thing we are measuring has gone "out of control." If I show this to a worker actively observing a piece of machinery, this image alone might spur the person to action. However, if I show it to the CEO, I'm liable to get a blank stare.
4. The dataset
The dataset we will use throughout the course will be the sales database of a fictitious UK clothing business known as Threads Ltd. The company operates across the island through a network of retailers and chain stores.
5. The dataset
From our dataset, the orders table stores information related to sales orders carried out by Threads Ltd. This includes the key financial information related to each order, such as the sales amount and the cost of goods sold - also known as COGS. This variable indicates the direct costs associated with making the clothes the company sells. We also have the returns table that details any orders that were returned after purchase due to errors or faults.
There are also several dimension tables that we will be working with. The products table contains information related to the items of clothing that are manufactured by Threads Ltd. The retailers table holds all relevant data to the retailers that purchase the items - essentially, the company's customers.
6. Our audience
Our initial audience will be the Head of Sales of the company, who has a good understanding of the high-level figures related to the company's performance and the products they sell.
However, the Head of Sales does wish to investigate in more detail what type of orders the company is processing and what products are being ordered.
The specific output requested will be tailored to her requirements as it is a special project she has asked you to perform.
7. Let's practice!
Now that we've learned a little bit about the data and our intended audience, let's dive into some exercises!