CommonLounge Archive

Designing Data Dashboards and Actions in Tableau

April 18, 2018

In this tutorial, we shall learn about designing Data Dashboards in Tableau. The topics being covered in this lesson are:

  • Creating a Scatter Plot
  • Creating an Area Chart & Learning About Highlighting
  • Contextualizing analysis results by creating Tableau dashboards
  • Incorporating charts and conditional formatting into dashboards using supporting information specific to a particular data set
  • Adding an Interactive Action Filter and applying Filters to Multiple Worksheets
  • Adding Interactive Action Highlighting

So, let’s begin.

Getting Started

Open the Tableau Public application.

Select Excel from Connections and add the Sample Superstore Data from the directory that stores your local saved version.

After the dataset connects, drag the Orders set to the center of the worksheet and wait for it to load.

At the bottom left, click on the icon that says “Sheet 1”.

Your blank worksheet will open up ready for visualization.


First, we are going to create a scatterplot. This is a commonly used chart to view the relation, if any exists, between two numerical fields. We will use the measures of Profit versus Discount to make our scatterplot here.

From the Measures panel on bottom left panel, drag the field Discount to the Columns section on top and the Profit field to the Rows section.

We can see only one point here. That is because all the data has been aggregated by default. To view the individual data points, select Analysis from the top header pane.

You will see that the second option Aggregate Measures show a tick mark indicating this default setting has been applied. Click on it to toggle the selection. The aggregation will be removed. Our viz now changes to:

Click on the Show Me button on the top right of the worksheet. We can see that Tableau has automatically rendered a Scatterplot.

As viewed in the notes below the Show Me pop-up, we need 2-4 Measures to make a scatterplot and 0 or more Dimensions. Our viz has 2 Measures; hence, the scatterplot is displayed automatically by Tableau.

Here, we observe that there is an inverse relationship between discount and profits and that there is high probability of a correlation between these two variables.

Moreover, as profit is usually a target variable or performance metric and we can see its linear relationship with discount, we can conduct linear regression analysis* to determine a more exact impact of discount on profits. This can be useful in forecasting profits in future time spans given a certain discount value.

*Note: Linear regression analysis is a basic and commonly used type of predictive analysis. Regression is used to determine predictor variables- (i.e. the variables that influence outcome), their significance and impact. For example, in our sales data, what factors would determine profit? Discount? Sales? Region? Category? By how much? Knowing these answer can help us predict profits given a certain amount of sales in a certain category and region with a particular discount.

Now, let us continue with our further exploration.

From the top left panel, select the Analytics pane.

Observe the Analytics pane that results on making this selection.

We can observe the rough trend visually on this plot — that profits dip with increase in discounts. We can create a trend line to highlight this observation.

Trend Line

A trend line can provide a statistical definition of the relationship between two numerical values. To add trend lines to a view, both axes must contain a field that can be interpreted as a number — by definition, that is always the case with a scatter plot. In our example, the trend line can showcase the relationship between Profit and Discount.

In the Analytics pane, under Model section, click on the Trend Line option. Drag it to the worksheet to view different options available. Drop it on the Linear option as follows:

The viz will now have a trend line as given below.

Do explore the other types of trend lines. For this chart, they will almost be identical. Now we shall try to discern the losses and profits by using the Color card. As learnt in previous lessons, drag Profit to the Colors Card from the Data pane.

The viz will change as below.

As learnt in the last tutorial, change the colors to range between Red for Loss and Green for Profit with the Stepped color option. Stepped color will ensure distinct color difference corresponding to how huge the profits and losses are.

The viz will now transform based on updated settings as:

It is clear here that we have too many points in the mid range that are profits and losses. Hence, we need to increase the number of stepped colors to make sure that profits and losses are not clubbed together. This is an important lesson in analysis and visualization: that several iterations are needed to understand the data, extract insights and make it presentable.

Change the number of stepped color to 6.

Our viz now looks better.

A few points do overlap on the trend line but it is more representative of revenue than the previous version. Save the viz to Tableau public as “SalesDashboard”. Your scatterplot is now complete as is the first plot for your first dashboard. Congratulations!

Line Chart

Now we shall create line charts. These work best with continuous variables like date-time. So, open a new worksheet using the icon next to Sheet 1 in the bottom panel.

The new worksheet is blank.

Click on the Show Me button to see what kind of variables are necessary to make a line chart.

As mentioned above, date times are usually best suited to line charts. So, we will use Order date for the date aspect. We have covered Time Series and Level of Detail in the previous tutorial and will apply the same concept here.

Add Order date to the Columns section. Set the Level of Detail to Month. Day is the most granular for this variable.

Add Discount to the Rows column. Our viz updates as follows.

Now click on the Show Me panel. We see that the second Line chart is auto-selected. Click on the first (leftmost) to get continuous lines.

Our viz looks much better now.

We can also explicitly highlight data points using the text icon on the top menu.

We can also accomplish this by using the Text card on Marks panel. Select the check box that says Show mark labels.

Our viz now display each data point with values.

Save the worksheet. It gets saved as a new tab on the same Public Tableau worksheet.

Note: Screenshot shows an additional tab on our sheet but should only have 2 on yours

Repeat these steps on a new worksheet but with Profit Measure instead of Discount. You should arrive at the following:

In this chart, we will highlight the profit-loss aspect using colors. As in the previous visualization (scatterplot), use Red-Green color with selections as below.

Our visualization now highlights profit and loss using divergent colors.

Area Charts

An area chart is a line chart where the area between the line and the axis is shaded with a color. These charts are typically used to represent accumulated totals over time and are the conventional way to display relative proportions of totals or percentage relationships. The area chart is a combination of a line graph and a stacked bar chart. By stacking the volume beneath the line, the chart shows the total of the fields as well as their relative size to each other.

In our example, it will be useful to study the relative sizing of discounts and Profits for all the Ship Modes. So, we shall create area charts utilizing these line charts we created. First, duplicate the two line charts using the menu controls as follows:

We should now have five sheets in all as follows:

Go to Sheet 2 (2) that is the duplicate of our Discount chart. Add Ship Mode to the Rows section.

Our viz will have four separate line charts.

Go to Show Me panel and select the first Area chart.

Our visualization is modified to display Area charts.

Using Text Labels icon, instantly add data point values to the chart.

Our viz will be as below.

Select the second line chart showing Profits (on Sheet 3(2)). Add Ship Mode to Rows.

From the Marks panel, select Area chart instead of Default.

Our visual is changed in appearance now.

Data Dashboards

Now, we shall incorporate all these charts into a dashboard.

Why are dashboards important?

Just like the dashboard on a plane or car, Tableau Data Dashboards provide an overview of the different metrics or performance indicators that you want to measure. They can provide snapshots of individual elements as well a summary of the entire system as a whole.

If you have ever seen a Twitter or LinkedIn or website Analytics dashboards, you’ve seen a data dashboard which gives you a summary of metrics important to that medium like visitors, connections/ followers, shares, likes, average statistics, performance in comparison to previous spans, best and least performing days and so on. This ability to provide both monitoring and understanding at both the elemental and holistic level is what any good dashboard aims to do.

So, let’s begin.

Rename the sheets with intuitive names and save.

Select the Dashboard option from the bottom panel as given here.

It opens up a new dashboard worksheet as below.

Change the layout size by selecting controls as displayed below.

On hovering over the Sheets menu, we observe the different worksheets we created.

Drag “Discount over Time” viz to the sheet.

Add viz called “Profit over Time” to the bottom portion of the sheet.

Dashboard is altered as given below. The reason why we are aligning side to side is that axes are different and will not be conducive to side-by-side comparison.

Add the Profit by Ship Mode and Discount by Ship Mode adjacent to the respective graphs.

Remove the Color Legend by clicking on the X symbol next to it.

Action Highlighting and Filters

Now we will select actions and highlighting options on our dashboard.

Tableau allows us to add context and interactivity to your data using actions. There are three kinds of actions in Tableau: Filter, Highlight, and URL actions (Tableau Desktop only).

Actions are useful to empower exploration by audiences, to enable viewing certain segments / variables on our dashboards or inspect selected values on our visuals to check for patterns or trends.

In our charts, we will see the effect of Ship Mode and Date Range on Profit and Discount. We have different methods called Actions to do this. We will use highlighting to see the effect of each Ship Mode and Filters to focus on date ranges of interest. Let’s start creating.

From the menu, select Dashboard → Actions as below.

A pop-up window appears as below.

Click on the Add Action > Button. Select Highlight from sub-menu.

This opens another window as below. We wish to see the details for every Ship Mode — we want to explore if there is any impact of Ship Mode on our metrics. We will accordingly use the options to highlight values in our dashboard as shown.

The rationale behind this is as follows — our dashboard currently shows the profits and discounts over time, by profit / loss and drilled down by Ship Mode. We add interactivity to view individual Ship modes by hovering on a single value of Ship Mode. This helps us see how a particular class of delivery attributed to revenue. It also shows how all transactions related to a specific Ship Mode fared over time, what kind of profits or loss it accrued and so on.

Use the selection controls as below to enable this.

We want to see the Hover effect on all the sheets included in our dashboard. Hence, we add the the four sheets to Source. So, if we hover on the Ship Mode on any of the plots, the action is instantiated.

We want to see the corresponding values on the other visuals as well. So, add the four sheets to Target sheet. Thus, the action is applied to our plots of choice.

Our viz now incorporates highlighting as shown below. When we hover on first class, the values exclusively for First class on all plots are highlighted.

The left two plots contain aggregated values; they are not drilled down to separate values for each Ship Mode. Hence, note how our hover action does not highlight values here even though we selected these two sheets as well.

Notice on the first graph, that + and – symbols appear on the bottom left while hovering. These will expand the date ranges as needed.

Click on the down arrow on the top right of this chart. Select Filter and Month from the Menu.

The filter is added for use but is applied only to this worksheet. Apply it to all worksheets using the menu option by clicking the filter (funnel symbol) next to the slider filter.

On changing the time frame to 2013-2015, all the four graphs change in the dashboard.

The highlighting of specific Ship Modes with the Hover feature still works too as below. Hover on Standard Class to view the corresponding values as below.

As observed here, our dashboard gives us some interesting insights.

  1. By being able to drill down to single elements, we can see that First Class had the highest profit and the greatest fluctuations while Standard class, with least profits also was the most stable source of revenue.
  2. We also get a holistic view of the company’s financial health. Profits show a steady upward trend with the period around Jan 2014 showing slight fluctuations for Same Day ship Mode and wide fluctuations for all the other Ship Modes. As seen on the Profit by Ship mode graph, we can see the differences between totals in each following intervals and how much the totals vary. These could be followed up to understand if it was due to extraneous factors like economy, mergers, executive changes etc. or more controllable factors like policy or pricing changes.

As can be seen here, dashboards help you monitor, analyze, and further improve analysis by providing conclusions and additional exploratory paths.

Congratulations! You can now apply different types of actions like Filters and Highlighting to add interactivity and conduct in-depth analysis on your datasets, with the advantage of viewing it visually and holistically using dashboards.

A few directions for analysis might include:

  • Exploring performance of segments over time.
  • Applying filters and highlight actions to see relative and individual performance of categories and sub categories.
  • Conducting an analysis of manufacturers to see performance and to infer if any manufacturer benefited or suffered from discounts.

Keep practicing and keep exploring.

© 2016-2022. All rights reserved.