In this video, I'm going to talk about chart types. Specifically, we're going to look at why you would choose one chart type over the other, and how to make that decision. So here on the screen, I have some data already pulled up, and are three most commonly used chart types. We have vertical bar, we have a pie chart, and a score chart.

Now, score charts kind of speak for themselves. They're, in this case, they're a very simple score calculated. In, they almost always need some sort of context. So, 2715 is the number. But there needs to be some context shared with it. Is it, , good? Is it bad? Is it expected? Is it measured against some KPI?

So it alone doesn't serve the purpose. It needs either some design or some other context through other methods to add to it. But I'm going to go ahead and delete that one. Um, and we'll focus on telling a story with the pie chart and telling a story with the vertical bar chart and when and where that should happen.

Pie charts can be useful, and here's some of the things to consider when using the pie chart. A pie chart represents 100%, I only got one pie, and that pie is sliced up into a bunch of different pieces, but in total, all of these values should add up and equal 100%. Because what I'm looking at is how much of each.

Slice makes up the portion of. So typically when you do this, you want to see I have two or three variables. Ideally, how much is each one of those variables that I'm looking at take up for the full 100%. Now, a few things about this chart that probably don't work very well. One is we have 1 2 3 4 5 6 7 different slices.

All the slices here pretty similar, except these get so small that even the values overlap. The other thing is we're looking at numbers here. We're not looking at actually percentages. And again, we're looking at a percentage that each takes up so we can change this chart a little bit to make it a little bit better.

Um, I guess if we change the data values just into percentages and remove the labels, that helps. I could probably, like one of the things I would suggest is you only again use two or three, possibly four, variables here. But when we do that, what happens is we're actually taking away some valuable data.

So in this case, I would recommend that we don't use the pie chart, because we're not going to be able to get this down to just two or three variables. And if I'm showing this data, um, month over month, and I'm running this report, you know, it's possible that some months, I'm going to have variables that come up that I want to represent here, so I can't cut it down into just two or three slices.

So, in this case, if we go over here to the vertical bar chart, this tells us the magnitude of each of the categories. We are not limited just to 100 percent on a vertical bar chart. I'm showing the count of each category, and that could go, , as many bars as I need. They're, the 100 percent is there's no stopping point, so I could have even more bars and each bar could have, an infinite height in some regards.

And so, in this case, because there's so many categories, I'm trying to get a magnitude of each category. I want the exact values. I would suggest using the vertical bar chart. Let's go to the next slide. Now, in this case, I have a similar thing where I have a few. One, two, three, four, five, six different channels, but in this case, we could see that and we know in our business that our primary channels that we promote our phone, our chat and our email, our walk in and self service and any other are not promoted.

They're not supported really. So I really only care about looking at my phone, my chat and my email. So in this case, What I could do to clean this up a little bit is move this down just to four channels. I could go in my data. I could say set other category to true. And then I'll just change the number of slices down to four.

And now I have what I want to see, which is our phone is taking up thirty three percent. Our chat is taking up thirty three percent and our email is taking up thirty three percent. And then all the leftover channels, which are not even things we care about, are just 0. 1. So again, ultimately, the number of variables I care about representing here, it's just four.

So a pie chart works out pretty well. And again, I want to see the proportion of the whole that they take up. Now, when we look at that same data over here, it does work. I can see the magnitude of each of those channels, and then I could see the cell service and walk in or, you know, very little, probably in this case, just errors in the data.

So in this case, this does work. But again, if I'm just wanting to see which channel, how the distribution of channels across those three, maybe in this case the pie chart works better than the bar chart. Hopefully that gives you a little bit to think about when you're trying to make a decision between using a pie chart, using some sort of bar chart.

In this case, the pie chart works. In this case, the bar chart works. And, again, the data often determines the design that you're going to end up using. Some things to think about as you're setting up your next visualization