Join Gini von Courter for an in-depth discussion in this video Create chart visualizations, part of Power BI Pro Essential Training.
- [Voiceover] Let's add another page to our report and do a little more work with some of the traditional charts that we're used to creating in Excel and therefore, that people are used to seeing so they don't require a lot of explanation. Charts are relatively easy to create. A little more difficult than they are in Excel because we're not simply looking at data and selecting it. Therefore, you need to know what your data elements are. This is a good time for me to make a pitch about the difference between learning Power BI and using Power BI.
This data set that we're using is real data that's been made anonymous, but it's actual data from stores in someone's store chain. And so we're working with data that isn't my data, it's not your data, but it's available and accessible to us. When you're working with your own data, something, I think, close to miraculous happens because as you start asking questions about your own data and creating reports and dashboards to express things you'd like to know about your data and things you'd like to represent about your data.
I believe the entire work of Power BI gets much easier. It certainly gets more intriguing. You've got some skin in the game, and there is something at stake, but also, you have some expertise. I'm not an expert about this particular data set, and neither are you, someone is, but... You will be an expert about the datas that you are working with. So if all of this feels a little bit remote still at this point, don't let that bother you. Like data modeling, like working with pivot tables, working in Power BI becomes easier the closer you are to the actual data that you're working with.
Let's create some charts. If we look, for example, at items, we have categories, family name, buyer. I'm gonna pitch category out here, and we'll get a list to begin with. It's a pretty short list, and it has the names of items on it. I'm going to simply make those a little bit more visible by kicking up the font a bit... so that we can all see them. But what I'd really like to do is create a chart of this data, and you'll notice that there are no values here.
The only items that we have are really lists. So I need to go find some numbers, and notice that we have some calculated numbers. This isn't transaction data, this is summary data. So if I would like to know, for example, the sales by category last year, let's go ahead and click sales. And we'll get total units. It was a table, it continues to be a table, but now, if we click, for example, the pie chart, we'll get a pie chart of that same data.
Let's do this in the opposite direction. I'm going to simply press delete and get rid of that chart. Let's start, instead, with the total units that we sold last year. Because I start with a number, I'm getting a chart. I can make it a pie, it will be amazingly like a round circle, but now when I add category, notice that it flushes out the chart that's already there. So if I'm choosing to measure its value, I'll get a chart. If I'm choosing to measure it's geospatial data, like city or postal code, I'll get a map.
If I choose labels, then I'm going to get a table. Begins to make sense, then, that I might want to choose first numbers if what I intend to do is create a chart, but it's so easy to change it. It's not really that big of a deal. Let's make this a little bit bigger. And if we wanted to, again, make our font larger on, for example, the title, here's my title. Font.
If I'd like to make my data labels larger... Here they are. That looks pretty good, not a problem. Now, let's add the total units this year. When I do, really nothing changes, and the reason is the same as in Excel. I can only show one data series in a pie chart, so at this point, let's go ahead and choose a different type of visualization, and this is my stacked column chart or a stacked bar chart, either one.
And if I wish, I could turn the legend off or on, notice up here at the top. I can modify my data colors as we did previously. I can turn data labels on or off and set their size. So I have lots of choices about what I might wish to do. I can go to the X axis and change the color, for example, of the font that's being used.
So I can get very granular about what I want to modify in a chart. If I'd like to reset anything, notice, revert to default, everything falls back to where it was. And if we wanted to look, for example, at the average sales price... Again, new visualization, either click in the canvas and then choose or drag and drop, and there's the average unit price, and if I want that by category, choose category.
And there's the average price of different items that are being sold and, again, back to modifying, for example, the title size... so that it's similar. And perhaps I wanna turn this one sideways. That's fine. Very easy to create charts. Now we have some other choices of charts as well, as well as our bar charts and our line charts, and our pie charts, our basic charts.
Combo charts, a couple of others that are, I think, very interesting and worth looking at. If we looked, for example, at our total units that were sold in this chart, I'm going to take out last year and leave just this year. This is total units this year by category, which we could represent as a pie, but an alternative to a pie is something that's called a tree map, and it's a lot like a pie, but instead of representing a pie, it represents boxes.
Each of these are what would be a slice, and you can look at their size relative to each other. This is a very common type of a data representation in Power BI, particularly when you have not a lot of numbers that are similar but some that are much larger because it's very flat and easy to see, but also, because you can represent hierarchical data here. So if we added buyers, for example, to this, now what I have is I actually have the items, but who we bought them from.
Let me make this a little bit bigger. Let me just move it out of the way. We're just gonna slide that over here so that I can make this one bigger because what we now have... Allow me to make this one smaller just so it doesn't get crazy. So what I now have is in each category, I have who the buyer was for each of those items. If I have data that can easily be represented in this type of a fashion, you can't do this with a pie chart.
I can do it with a donut chart. I can do it with a stacked column or bar chart, but this actually lets me see first the category and how those compare to each other, and then within a category, we can see, for example, we only have four buyers here for groceries, but when we look at the women's area, we have nine or 10 different buyers who are available. I could then add simply a list of buyers, and if we were looking at a particular category, for example, shoes, here is the tree map for shoes, and there is the list of buyers for shoes.
If I choose groceries... Pretty slick. So this allows us to look at categories and buyers, and therefore, I'm going to rename this page for categories... and buyers. Just like that. And here's a tree map visualization as well as a more traditional bar chart visualization to represent our category and buyer info.
- Signing up for Power BI Pro
- Connecting to data sources
- Uploading data such as CSV and XLS files
- Creating reports, visualizations, charts, and maps
- Filtering, sorting, copying, and pasting visualizations
- Downloading custom visuals from the gallery
- Modifying existing reports
- Creating and managing data dashboards
- Querying data with Power BI Q&A and Microsoft Cortana
- Sharing report and dashboards
- Using Power BI Desktop and mobile apps