Advanced Visualizations

In Count there are times when you'll want to do more with your visuals than the templated options allow. This is where the custom visual builder comes in. To access this I like to start with a template and then click on this custom tab. Here we have access to the full suite of visualization options and channels. Let's start with marks. Most templated options are made up of single marks. You can see here I have an area mark. I now have the option to change this though and you can see the shapes that we have in this drop down. I could, for example, change this into a bar chart. Marks are the building blocks of any custom chart and we can add and layer these to to add more depth to your visuals. So to add a new mark I can either use the plus button here or I can duplicate an existing mark which is often a quicker way to do it. So now you can see that I have two bar marks and they are set up in exactly the same way so they are just sitting on top of each other. But I can change one of these. Let's choose line. If I change the color to make this more obvious you can now see that I have a line following the same data series. To customize this further I could change what the line is depicting. Let's change it from sales to profit. By selecting either of my marks, I can go in and customize them independently. So this looks good. They're currently both using the same axis. I could rename this to make that more obvious or another option I have if I go into the options for my profit field under display I can toggle on using a secondary axis and perhaps I will adjust the format as well. The last thing to know about marks is that they are layered on your visual depending on the order that they appear in this design panel and you can simply drag these if you want to change the order of that layering. The other thing you'll notice in this custom view is that you have access to the full suite of channels that you're able to encode. One that you may not have seen before is detail and this allows you to add additional information to your tooltips. I can drag a field over, it does not visually display but when I'm hovering over I can now see category and my tooltip. Next let's look at facets. These are great when you want to see the same chart broken down into different categories. For example let's take region and put that into facet rows. If I make this bigger here we can see that the same chart is repeated for each of the four regions within my dataset. We could go one step further and perhaps I'll add region to color as well just to distinguish these even further. Facets is a great option. If you were keen to compare exactly the same charts between each other. So these share the same axis and the same scales. If you need a little bit more control than that, then we do have another option for layering charts using subplots. It's useful to consider this grid layout when you're planning subplots. By default each chart you create has some subplot coordinates x and y which default to one. If you want to add a new mark as a subplot you can have it appear below your original by changing the y coordinate to two or appearing to the right of your original by changing the x coordinate to two. Let's see what this looks like in practice. I'm going to duplicate my bar mark and to assign this one to a new subplot I simply click on the three dots next to the type and here you will see the options. I'm going to change y to two and now we see that it is appearing below my original chart. You can see that these charts share one common axis, but the remaining axes are entirely customizable. For example, instead of showing the same thing, let's change sales to profit. You can see now that we have two independent axes here. I could independently change the color. This gives you a little bit more flexibility if you need it compared to facets. Next we'll look at pivoting. Pivoting takes multiple columns and reshapes them into a single long column which makes it much easier to compare measures side by side in a visual. Let's look at what this looks like in a pivot table. Here we can see that my three columns with values are now within a single longer column called measure names with the associated values next to them. So let's look at why this is useful. If I wanted to plot these three measures on a visual, I could typically do that through using different marks. You have a line chart. I'm gonna click into custom. I'm going to duplicate the marks and I'm going to customize each one. So this is one solution to be able to combine this data into a visual but let's look at the option if we were to pivot our data. Let's start with the same line visual that I started with before and this time I'm going to go to the custom tab. If I scroll down we can see the pivot area here and I'm simply going to add my columns to this. The initial hasn't changed yet. What you will notice is that in the custom view we have the addition of fields called measure names and measure values. If we look back at our pivot table we can see that the measure names relate to which column we're referring to and the measure values refer to the associated values. So we can now use those in our visual. Going to map our values to the y axis, and I'm going to map our measure names into the color channel. And here we can see that just by using a single mark, we've been able to create the same chart and compare these three. So that can be a much more efficient way to build these types of charts when the data that you're comparing share the same axes and scales. Another thing this allows us to do for example is if I switch this to a bar type it allows us to create stacked bar charts very easily. Just note that with this mark type if you click on the three dots beside it you also have the option to disable stacking if you'd like these to appear beside each other. Next we'll look at some of the different ways that you can aggregate and perform calculations on your data within visuals. By selecting the three dots next to a field you will see the preset aggregations that you're perhaps already familiar with. Let's have a look at bin, this allows you to group numeric values into buckets that will allow you to simplify your chart. By selecting this it will automatically bucket my values depending on the range that I have. I can also adjust these manually. The other thing that I may want to do is that I'm currently displaying this data as a continuous type. If I want this to appear as a histogram with discrete types I can simply change it here. Next we'll look at table calculations which allow you to transform your data after it's already been aggregated in your visual. Using a pivot table for this example, I'll come to the logic option for my values and here I see the table calculation option. I can choose the transformation that I would like to apply. Here I'll go for percent of total and you'll see I also have the option to provide context to that. Let's have a look at table. So this has calculated the percentages across the entire table, all columns and all rows. If we go back in and change this to table down this is now showing the percentages just for columns. You can explore the definitions of all those contexts within our documentation. It's also possible to apply custom calculations in visuals when you need more flexibility than the built in functions. In this example my scatterplot looks at sales and profit but if I wanted to create a new metric here I can click on the three dots and choose edit expression. Here, I'm simply gonna specify profit over sales. To commit this, I hold down shift and return. Now you can see my access has been updated to reflect this calculation. And by clicking in display I could give this a more meaningful name on the axis. Adding your own calculations doesn't stop at simple arithmetic, you can also use functions as well. Let's pull another field into the color channel. Here I'm showing a color gradient based on the discount applied to these products but perhaps I just want to display whether products had a discount or not. I'm going to edit this expression this time using a shortcut by double clicking on discount. I'm going to use a case when statement. Here we can see we now just have two colors represented and this is also displayed in our legend. We can also adjust this display too by giving the legend a meaningful name and we could change the color palette here of the two categories that we have. If you'd like to explore the full list of functions and expressions that are available to you click on this link to the docs. If you find yourself repeating the same calculations across charts it's really easy to create a shared catalog so this logic is reusable to you and others and it's consistent across your organization. You can find out more information about this in our learning materials on Kite metrics. Now we've covered the full functionality of visuals in count. I'd encourage you to explore this canvas, every visual under the sun, for inspiration on how well these features can be combined to create all these different charts. You can also access these from the visual section of our template menu, and this is really useful because it allows you to add these charts into your own Canvas so you can look at the underlying data and really see how each of these charts have been made. Okay. That's it. I hope this has helped you be much more creative and confident with your own visualizations in count. Thank you.