The more care you take to create your reports, the more the user will get out of them. I’d like to think about the information the user can gain in the first five seconds of seeing a graph. If they’re still trying to work out what on earth the graph is showing them, then there’s definitely scope for improving the design. So essentially during this presentation, we’ll look at three key techniques. After a quick introduction, the first technique will be enhancing a default visual. By that, I mean taking a graph that you can create out of the box from SAS Visual Analytics and exploring all the available options and settings to make sure the graph really shows what you want it to. The second thing we’ll look at is the SAS Graph Builder. It’s an application that allows you to define your own custom graph templates and add them to the list in the Object’s tab to be used when you’re creating reports. The last thing we look at is the Data-driven Content object. It’s a really powerful object and allows you to embed third-party visualizations inside your SAS visual analytics report. So that’s what we’ll cover. We’ll start off with the introduction. When you first create a report in SAS Visual Analytics, the type of the object will show you around 40 different visuals.
Absolutely loads of choice for the graphs that you use to create your reports. A few of the most common ones are included, including pie charts, heat maps, histograms. As discussed, it’s important to create the right graph. But there are so many different types of data that you could be working with, there is no single visualization to present all of your data. Before we have a look at an example of that, the case study we’re using throughout this presentation is based on a UK gym company looking both at the total number of members that each gym currently has and also the targets that have been set for the number of members in each gym. So as we look at the examples, we refer to all of those. OK, so we’ll first look at what makes a graph good. And then we’ll look at an example that explores different options and settings in SAS Visual Analytics for a graph that we’ll create. So to start off with, I’ve got two graphs on the screen, graph one, graph two. I’m going to go through a couple of questions, and I’d like you to think about which graph you’d use to try and answer these questions. Which gym has the most members? The second question is which gym has more members, Hackney Town or Islington? Hopefully, you picked graph two. Both these graphs are showing exactly the same data, but graph one, although colorful, is quite hard to get any meaning from. You can’t tell the differences between each of the gyms quickly. Graph two, on the other hand, shows each of the bars. The length determines the number of members, and the number is also printed on it. It shows that there are completely different ways of presenting the same data, but not all are good. So whenever you’re creating a graph, it’s really good to consider how you’re presenting it.
We’ll now look at an example in SAS visual analytics of a graph that we’ve created, the problems that it shows, and how to overcome those just with the settings and options available. OK, so the following graph is a targeted bar chart. It’s showing each gym across the bottom. The bars are showing the total members at each gym, and the black markers on each of the bars are showing the targets that have been set for each gym. Now the first problem that I decided with this graph is that the gym names are not fully visible. You could hover over and get some more information. Visual analytics is interactive. But really, you’re trying to let the user see as much as they can right from the beginning, so that’s the first thing we’ll address. A second thing is you cannot easily see how many gyms are below their targets. You could go through and study each of the bars, but there are lots there so the information is getting lost. The last one is it’s not easy to see which gyms are performing the worst. That is, which gyms are furthest from meeting their targets if they are not meeting their targets. So let’s have a look at what we can do in SAS Visual Analytics to overcome some of those things. Things to consider are generally found on the right-hand side when you click on a graph, including Options, Rules, Filters, and Ranks. So Options are generally used for controlling the properties of the axes, the titles, the legends, the overall look and feel that helps the user understand what they’re seeing.
The Rules, also known as display rules, allow you to create logic-based rules to dynamically color parts of your graph. Styles should also be used. They sit under the Options tab, but they can set standard themes throughout your reports to make consistency and make that a way of allowing users to see what’s happening straight away. The last couple of things to consider are Filters and Ranks. Filters allow you to create logic-based rules to control how much data is being presented in the graph and Ranks are similar. You’re controlling how much data is shown in the graph, but instead of based on some logic it’s simply a top or bottom count. So, for example, if we’re showing a total number of members, we could set a rank to show the gyms with the lowest number of members. And we’d pick how many we wanted to see, so 10 or 15. So let’s have a look at a few of those in action for our targeted bar chart. First of all, the Options. The first one I highlighted is setting the title. If the title’s clear and it explains exactly what the graph is showing, users will understand quite quickly what they’re looking at. We’ll also set the x-axis option to rotate the gym names. You can create as many display rows as you want. I’ve created two different ones here based on a column called Difference. The column called Difference is a calculation between the total number of members taking away the target we set them.
So if the difference is a negative number, the gym has not met its target. if it’s a positive number, they have met their target. So here we’ve set one Display rule. If the difference is positive, the bar will be gray. If it’s negative, we’ll make it red so it quickly stands out. And last but not least, we’ll add a rank. We’ll limit the amount of information shown in the graph to only those gyms that really aren’t performing very well. So we’ve created a rank on gym name looking at the bottom count. We want to see 15 gyms, and we’ll evaluate what counts as a bottom by the columns. And so the most negative values will be shown. And the result of that is this graph. So we’ve got the title. We’ve got each of the bars for the 15 lowest performing gyms, those gyms that have the biggest negative or difference between their targets and the amount they have at the moment. Across the bottom are all the gym names quite clearly presented, and the color-coding on the bars of those that aren’t meeting their targets. To summarize, whenever you’re using SAS Visual Analytics you do have lots of options for the graph that you use.
The first thing is to pick the graph that is most suitable for the data you’re working with. Once you’ve done that, there are absolutely tons of settings and options that you can use to customize the output that you’re creating. Always consider these steps whenever you’re creating a new graph, regardless of whether you’ve decided it’s not good enough or not. Always explore the options that we’ve discussed. OK. So next up is the second technique that we’ll cover to enhance how your reports look. This one’s using a slightly different application that comes with your SAS Viya environment called the SAS Graph Builder. This technique’s about creating your own graph templates and adding them to the list of objects. As you can see in the screenshot, the standard list of graphs that you can create is there. And the highlighted one in blue says line chart with time flag. We’ve created that in the Graph Builder and then brought it into the object so that we can use it as a normal graph.