Creating visuals

- [Instructor] Worksheets are the foundation of our dashboards, and so to make our data come alive, we want to make it visual using Tableau's charts and graphs. So now that we've connected to and imported our data, we can see all of our Dimensions over on the left-hand side, so the Dimensions are the kind of category levels, and then the Measures, typically kind of numerical, what we're going to be using to evaluate pieces. So we can use the Show Me button, or as you get started with this, you can use the Show Me button to see a lot of the different default chart options.

Now, different charts are going to become available as you click on different values. If I start in the Data_Revised and I click on Country, you'll see one of the first things that's pops up is a map because it's automatically associating the Country here with the map. As you start to pick different values, so if I pick Series Name and then hold down Control and click on my Mobile Phone Subscriptions, now I've got different chart options available over on the right-hand side. So the chart types are really based on a set number of dimensions or data types, and you can see that by hovering over the different values.

So to get a tree map, you need one more dimensions, and one or two measures. So this gives you an indicator of what you might need for each of the different charts if you were trying to pursue a particular chart type, or if you wanted to explore one. So there's a lot to choose from, bar charts, pie charts, maps, lines, area charts, et cetera, and so when designing a dashboard, we really need to remember our following best practices. First of all we want to keep it visual, so whenever possible we want to use charts instead of tables. Data tables have their place, but take advantage of the charts.

The keep it visual best practice is easy to follow here but the real danger is to over-design the dashboard and make it too complex or confusing for the viewer. So we want to make it easy to use and easy to read. This also means maybe limiting it to three to five charts. Too much information can be just as dangerous, so we want to focus on really the critical charts for our dashboard. So each chart should also be able to stand on its own, allowing a viewer to read and understand the information.

We really want to keep our viewers at the forefront of our design decisions, and there is certainly a lot of creativity that can be put into the dashboards, and you may want to wow people with a look, but if it's too hard to read or analyze the data, the dashboard won't be effective long-term. We also want to make sure to pick appropriate chart types, so though not an absolute, it's good to study up on when to use various chart types, like bar graphs for categories, lines for time series, pie for percentage comparisons, scatter plots for relationships.

Tableau actually has a really great resource that they've put together on when and how to use various chart types. The link for the Tableau resource is down here at the bottom, so if we jump over and take a look at it, and this is included in your exercise files. As you'll scroll through, this is really which chart or graph is right for you, and so it helps you walk through, and you can see on the top kind of 13 different chart types that are available. There's certainly a lot of resources out there around which chart type to use when, but kind of a nice guide or reminder about when certain charts might be the best for the data or for the type of message you're trying to display.

So certainly a resource that you may want to take a look through to help understand which charts should be used with which type of data. Then finally, really as you're working with the dashboards, the next big key is really just to experiment. So find your style through practice, and as we start going down the road of designing our dashboard here, we're going to pick certain chart types, but we may want to experiment a little bit too. Tableau has so many options, that the best way to learn is just through practice and experimentation, so can we add color or size to give more information? Can we modify the chart types? Can we use dual axes or overlapping charts? But how can we just play with our charts to really make it visually interesting? So there's lots of options for visualizing our data in Tableau.

I mean that is the beauty of the program, but we want to make sure to keep our best practices in mind as we start to experiment and work with charts.