PL-300 – Section 6: Part 1 Level 5: Other visualizations Part 2

59. Pie Charts and Donut Charts

In this video, we’re going to have a look at pie charts and donut charts, and they’re not that difficult. If I’d add in a pie chart. Let’s say, I wanted to see what the total sales volume was. Okay, so 100% of all sales were made. That doesn’t tell me anything. So, now let’s divide it into the region name, and I’m going to add region name into the details section. And here we can see, we have got these six different regions and we can see what percentage of the overall sales were in Greater Manchester, West Midlands, West Yorkshire, and so forth.

Now, there’s not too much more in this until I add something else. Suppose, I wanted to subdivide each particular pie. Each particular portion. Say, the Greater Manchester. I wanted to know what the sales were in quarter one, quarter two, quarter three, and quarter four, but I want the colours to remain as each individual region name.

So, what I’m going to do is drag region name into the legend section. Nothing visible has actually changed. And now I’m going to go into the date hierarchy and drag quarter into details. And now you can see that each of these pie segments have been divided up into four. So, you can see quarter one, quarter two, quarter three, quarter four, for each individual region name. So, if you’re going to do this, make sure you put the legend on.

Now, the difference between a pie chart and a donut chart is simply whether there is a hole in the middle or not. So, you can change the data colours individually. So, maybe, I wanted Greater Manchester to be in a purple. Absolutely fine. You can change the labels so that, for instance, instead of seeing the category. So 1, 2, 3, 4. You could have the data value, or the percentage of the total, or some combination thereof, or of course you could just switch them off if it’s fairly obvious what each segment is for. And that’s about it, and it’s a bit disappointing in terms of the functionality. For instance, what if you had 20 items? So, I’ll change this from region name to date, year. It gets very crowded very quickly.

Now, in some other places, some other programmes, you could say “I want the top 10 to be shown separately and everything else to be grouped together.” That is not yet possible in Power BI. Hopefully, they will be expanding this so that the standard Power BI pie and donut could have some greater flexibility into it. But for now, if you have got some data which adds up to 100% and you want a visualisation of how each component makes up the total, then please have a look at the pie chart or the donut chart.

So, just a quick look at some of the other properties and they largely revolve around this details label. So, in addition to what you show, you can also say where see it. So outside, inside, or a preference. Prefer outside, Prefer inside. So, the prefer means that not everything will be outside. Then you can format the font. You can say whether a background is on or not. So, here you can see that the value here is shown with a background, but I can turn it off if I so wish. You can also display units in none, or thousands, or millions, and you can also say, “Well, I want this to one decimal place.” So, we’re having 173.3 as opposed to just 170, and you could also say the same thing for the percentage. I just want it to one decimal place. So, that is all in the detailed labels. And then in rotation, that just shows where we start with all of the colours.

In addition to these, you have the standard properties as always. So, you can add a title, and you can add backgrounds, and visual borders, and shadows, for instance.

So, these are some of the properties that you have got with pie and donut charts.

60. Tree maps

Now, an alternate way of displaying values which are 100% of a total value is by using the tree map.

So, if I duplicate this page and call it tree map we can just change this visualisation to a tree map over here, and you can see that we have the entire visualisation taken up by each of the region names and we can also drill down to show all of the quarters.

Equally, we could drag the date from the grouping down to the details to get the same effect, except there is a difference. While we have a look at expanding to the next level, all of these different boxes have got different colours. Whereas if it’s put into the details section then they all have the same colour, so greater Manchester, the four different quarters are all grouped together quite nicely. So, they all add up to the same thing, even if the details are removed, it’s just breaking this down into more detail.

And again, this sort of thing, like the other charts we’ve been showing, might be really good for adding drill throughs. So, imagine clicking on any one of these and getting to a visualisation for that particular area. So, you can drill through to something, a different page that you have set up.

So, tree maps, fairly straight forward. We have the grouping, only the top grouping is showing up first and then you can go further down, then you have details if you want to subdivide while still keeping the same colour. And then we have the sales, the volume, the values at the end.

Now, there’s nothing new in the formatting, all the formatting that we’ve seen previously are included in this particular visualisation. So, very little new to learn. So, just an alternate way of visualising the 100% instead of being in a circle, you can have it in a square, and that is the tree map.

61. Funnel Charts

Now, the final of these other visualisations that I want to show, before we go on to mapping and measuring performance using KPI gauges and cards, is a funnel chart.

Now, a funnel chart shows percentages of the total, so rather like what we’ve had here from top to bottom. So, if I duplicate this and change this visualisation into a funnel chart, you can see that we have Greater Manchester having the greater sales volume, and then it getting progressively, well, smaller is the idea if this is a true funnel. So, if, for instance, instead of having the region name, we had the date, and specifically, the year, then you could imagine this showing figures, perhaps, going progressively out and then getting very much smaller. Funnel charts are good, if you’ve got sequential data, perhaps, you’re going through various stages, and you want to see how many hours are being spent by labour, how many hours are being spent by plumbing, how many hours are being spent by decorating, if you’re building a house. It’s really good, if the volume is going down each time.

So, if you went from 2007 to 2012, that could create quite a nice funnel, because it shows the largest value at the top and then going further down. It can also reveal bottlenecks in processes. So, imagine this was showing the number of hours worked in any particular trade, and we have more hours worked in the middle trade than in all the others, that could show that that middle trade could be a bottleneck.

So, anything which is a linear sequence is a good fit for the funnel charts. It could show, for instance, how much volume of a project you’ve got, as it goes through various stages, how much wastage there is at each time.

There’s only one new section in the formatting, the conversion rate label. So, what the conversion rate is, is showing where you were right at the beginning, and then where you were the previous, all the way down to the end. The conversion rate labels are the things right at the top, the 100% and the 41% in this case. So, if I switch them off, then you can see that they disappear. So these are funnel charts.

62. Adding Marketplace Visualization (Import a Custom Visual)

Now, even though these are the last of the visualisations apart from maps and measuring performance, there are additional visualisations that you can download.

If you click on this dot-dot-dot (…), import a custom visual, you can import from the marketplace. So, this will show you lots of additional visualisations that you can use. Now, most of them, perhaps, a bit more esoteric, some of these are building on the original Power BIs. So, for instance, if I look for the word “doughnut”, which can be spelled two different ways, with the U-G-H or without. You can see that some people have spent time looking at problems with the current Power BI, and are trying to find a way to expand on its functionality. So, in each case you can have a look at the various graphics, look at the description, see if this is something you want installing. And if you do want it installing, then all you have to do is click on the add.

Bear in mind, a lot of these are additional paid for options. So, click here to buy the advanced drill down donut chart. And you can see, also the reviews. Whether it is actually good, other people have found it useful. So, it can be worth just having a look down, seeing if there’s anything here that’s of particular interest to you and just downloading a having a play. So, it’s best for those that don’t require additional purchase but, sometimes the premium things may have exactly what you want. So, just have a look down, see if there’s something you want, and press the add button.

So, here we’ve just added a visualisation, and now we can just add it like any other visualisation and we can just use it as we see fit. So, I’ll add sales volume to the X, add inflation to the Y, and here we can see the scatter graph going all the way down to the data level it looks like.

And then you can explore the various different options that it gives you and see how you can customise it to make best use.

If you find that you’ve downloaded one that you don’t like, then you can always remove it. So, click on the dot-dot-dot (…), delete a custom visualisation. I’m going to delete the histogram with points. Yes, delete, and now it’s no longer included in my Power BI.

So, if you want to import a visualisation from the marketplace just click on the dot-dot-dot (…) in the visualisations panes, find it, and click add.