Chapter 7. Telling a Data Story with Dashboards

As you engage in data discovery and analysis, you will create numerous data visualizations. Each of these visualizations gives a snapshot of a story within the data. Each view into the data answers one or maybe a couple of questions. At times, the discovery and analysis are enough to make a key decision, and the cycle is complete. Many times, however, you will need to bring these visualizations together so that they communicate a comprehensive story to your intended audience.

Tableau allows you to bring together related data visualizations into a single dashboard. This dashboard could be a static view of various aspects of the data or a fully interactive environment that allows users to dynamically filter, drill down, and interact with data visualizations.

This chapter will cover the following topics:

• Objectives of dashboards

• Examples of dashboards

• Interactivity with actions

• Story points

We'll take a look at most of these concepts in the context of...

Dashboard objectives

Every dashboard seeks to tell a story by giving a clear picture of a certain set of information. Before designing a dashboard, you should understand what story the data tells. How you tell the story will depend on numerous factors, such as your audience, the way the audience will access the dashboard, and what response you want to elicit from your audience.

Stephen Few defines a dashboard as a "visual display of the most important information needed to achieve one or more objectives; consolidated and arranged on a single screen so the information can be monitored at a glance." This definition is helpful because it places some key boundaries around the data story and the way we will seek to tell it in Tableau. In general, your story should follow these guidelines:

• The story must focus on the most important information. Anything that does not communicate or support the main story should be excluded. You may include this information in other dashboards.

• The story that you...

Example – is least profitable always unprofitable?

Let's say you've been tasked with helping management for a superstore chain to find which items are the least profitable. The management feels that most of the least profitable items should be eliminated from the inventory. However, as you've done your analysis, you've discovered that certain items, while not profitable overall, have made profit at times in various locations. Your primary objective is to lead management through the discovery of the least profitable items and then guide them through a simple analysis that identifies whether an item has always been unprofitable.

Building the views

Let's start by creating the individual views that will comprise your dashboard:

1. Create a bar chart showing the profit by category. Sort the categories in a descending order by the sum of the profit.

2. Add the Department field to Filters and show a quick filter. You can do this by right-clicking on the Department field and selecting Show Quick Filter or...

How actions work

You saw a few examples of actions used in dashboards. We'll now consider some details of how actions work in Tableau.

Tableau supports three kinds of actions:

• Filter actions

• Highlight actions

• URL actions

Certain actions are automatically generated by Tableau based on shortcuts. For example, you can select Use as Filter from the drop-down menu of a view on a dashboard, resulting in an automatically generated filter action. Enabling highlighting using the button on a discrete color legend or on the toolbar will automatically generate a highlight action.

You can create or edit dashboard actions by navigating to Dashboard | Actions from the menu. Let's consider the details of each type of action.

Filter actions

Filter actions are defined by a source sheet(s) that passes one or more dimensional values as filters to target sheets upon an action. Remember that every mark on a sheet is defined by the unique intersection of dimensional values. When an action occurs involving one or more...

Example – a regional scorecard

We'll consider another example dashboard that demonstrates slightly different objectives. Let's say everyone in the organization has agreed upon a key performance indicator of the profit ratio. Furthermore, everyone agrees that the cutoff between good and poor profit ratio is 15 percent but would like to have the option of adjusting the value dynamically to see whether other targets would be better.

Consider the following dashboard:

This dashboard allows your audience to very quickly evaluate the performance of each region. Additionally, it is easy to see the historical performance evaluated by the standard of the KPI. The dashboard provides some interactivity with the KPI parameter and tooltips. Additional drilldown into other dashboards or views could be provided if desired. If this view were published on Tableau Server, it is not unreasonable to think that regional managers might subscribe to the view and receive a daily e-mail containing an up-to-date image...

Story points

The story points feature allows you to tell a story using interactive snapshots of dashboards and views. The snapshots become points in a story. This allows you to construct a guided narrative or even an entire presentation.

Let's consider an example in which story points might be useful. Executive managers are pleased with the regional scorecard dashboard you developed. Now, they want you to make a presentation to the board and highlight some specific issues for the East region. With minimal effort, you can transform even a simple scorecard into an entire story, following steps such as these:

1. First, we'll extend the scorecard with one additional view. Create a view similar to this one by double-clicking on the State field, adding Profit Ratio KPI to color, and including Profit Ratio on the Detail level of the Marks card. If needed, edit the color legend to use a Blue / Orange palette. If desired, increase the washout of the map to 100% (from the menu, navigate to Map | Options...

Summary

When you are ready to share your discovery and analysis, you will most likely use dashboards to relate the story to your audience. The way in which you tell the story will depend on your objectives as well as your audience and the mode of delivery. Using a combination of views, objects, parameters, quick filters, and legends, you can create an incredible framework to tell a data story. By introducing actions and interactivity, you can invite your audience to participate. Story points will allow you to bring together many snapshots of dashboards and views to craft and present entire narratives.

Next, we'll turn our attention to some deeper analysis with trends, distributions, and forecasting.