Tableau

 

Tableau Worksheet 

 Tableau worksheets are powerful analytical tools, but the real utility of this

application is in being able to share the analysis with other people in your

organization. An executive overview such as the one shown in Figure 1‑8 not

only breaks down sales by state, but it also enables the viewer to see how

different customer segments and product categories are performing. In addition,

the overview graphically displays profitability using different colors.

In Tableau terminology, this type of display that combines information from

more than one sheet is called a dashboard.

Tableau can use many different types of data sources, ranging from text

and Excel to all the best databases in the world.

The Tableau worksheet includes a number of elements that you’ll use as you

build your analysis. These include the following:

1) Data pane: This is the area that appears along the left side of the worksheet (in the Side bar) and contains two sections: one labeled Dimensions and one labeled Measures. These sections hold the fields that you can add to the work area to perform the analysis. The Side bar also has a tab labeled Analytics that’s used to add things like trend lines to a visualization.

2) Shelves and cards: These are the areas in the workspace with names like Pages, Filters, Marks, and so on where you drag fields from the Data pane to produce a visualization. Your visualization will change depending on where you drop a field, so later chapters will provide much more detailed information about using shelves and cards.

Understanding Tableau worksheets

Tableau has three different types of pages that you can use to create and present your data analysis results. These include worksheets, dashboards, and stories. 

You need to know the following:

 You use worksheets to create visualizations. 

You use dashboards to combine two or more worksheets that you want to share.

You use stories as a means of stepping people through worksheets and dashboards with commentary to guide them through your analysis.

Getting to Know the Tableau Desktop Environment

Getting to know Tableau Desktop means getting to know several different pages and workspaces. What you see in Tableau will depend on what you are trying to accomplish. Rather than presenting a cluttered workspace overloaded with controls and dialogue boxes, Tableau provides visual cues to suit the task at hand. 

Looking at the Tableau Start page

Tableau works with all kinds of source data that can be located almost anywhere as long as it’s accessible to the user. That source could be an Excel file, a text file, or a database. To do anything in Tableau, then, you first have to specify your data source or sources. That’s why, when you first start Tableau, you see the Start page shown in Figure

Tableau works using your existing data, so you must have access to some type of source data first. The page contains three distinct sections: 

Connect: You use this section when you want to start a new data connection in your Tableau workbook. Tableau Desktop has two editions, and the type of edition you’re using will change the options you have to choose from. Tableau Desktop Personal Edition is for connections to file- based data sources like Excel and Microsoft Access, and also includes several cloud-based data sources. Tableau Desktop Professional Edition allows you to go beyond the files and additionally connect to databases hosted on servers. 

Open: You use this section to open and continue working on an existing Tableau workbook. The existing workbook can be one that you’ve created or one of the samples that Tableau Software provides for training purposes. 

Discover: This section gives you easy access to training resources as well as news about Tableau.

 

Working with Dimensions and Measures

You’ve probably noticed that Tableau separates data fields into dimensions and measures,  It’s useful to understand how Tableau decides what fields to place in each area.

Understanding dimensions

Tableau treats any field that contains qualitative, categorical information such as text or dates as dimensions. These types of fields typically produce labels when you add them to the Rows or Columns shelves in a view. Dimensions enable you to provide detail in a view and to effectively slice or categorize your data.

Understanding measures

Measures are fields that contain quantitative (or numeric) values that you can do math on (sum, count, and so on). These types of fields typically produce the axes on a chart and are the numbers we use to evaluate whether results are good or bad. As a rule of thumb, most measures are numbers and most dimensions are non-numeric. Think about it in these terms. If you do math on it, it is probably a measure. If you use it to slice the data, it is a dimension. However, in some cases, a number may be a dimension. For example, while an order id may be a number, would you ever add up your order #’s or take an average of order #? You may, however, look at the Total Sales amount by order, so using the order id as a dimension would be preferred.

Modifying Your View

Tableau gives you a number of different tools to help you modify the view of the visualization. We take a quick look at a couple of them here. 

Transpose: This button swaps the position of the items on the Columns and Rows shelves. By clicking this button, you can quickly switch between horizontal and vertical bar charts, for example.  

Sort Descending: This button sorts a Dimension list in descending order.

Sort Ascending: This button sorts a Dimension list in ascending order.

Show Me: This displays the Show Me palette so that you can quickly choose different chart types.

 Understanding Data Sources

Modern companies live on data. They gather data on everything from inventory costs to labor costs to the smallest details  involving sales. All this raw data can then be transformed, aggregated, and analyzed into submission to create useful business information that can help drive competitive decision making. But before any of the data can be analyzed, it needs to be stored in an accessible and useful form. Now we take a look at what this means.

Considering how data is stored 

To actually use or analyze data, the data needs to be stored in a standardized format. In the days before computers, this typically meant writing everything down in a ledger. The ledger contained a number of columns that were used for specific purposes, such as the date of the transaction, the type of transaction, the amount of money involved, names of the people  involved, and other various details. All of this information was written in by hand, but the bookkeepers always followed the sameformat so that the information could be more easily understood. When computers came along, it quickly became clear that the old handwritten ledger could be replaced by a computerized database. What was also clear is the fact that the database needed to have a formal structure similar to that of the old-time ledger, because this formal structure made it possible for the computer to process the data. 

One very important thing to remember about databases is that they all have a defined structure.

Using file-based data sources

Unfortunately, people sometimes use tools to do jobs that aren’t totally appropriate for a given task. There is the old saying, “if all you have is a hammer, everything looks like a nail.” Using file-based sources like Excel instead of a database is an easy option for many of us, but needs to be done with care. Tableau will read the first few values of each column in the file and will determine a default data type. However, when connecting to a database, Tableau can pick up the definitions of fields from that database,  taking it more likely for your fields to be consistent. One of the reasons that people like to use file-based sources like Excel as a database is because they are so flexible. People don’t like to be told that they have to enter a valid date or other specific information that may not readily be at hand. Or, a user might decide that she would like to remove a column or use a different name for an existing column. Either way, if you’ve created a visualization in Tableau based upon the existing structure (and  hoping for valid data), your analysis could mysteriously stop functioning properly. As with any data source, be aware that you need consistency in your structure and keep an eye out for errors.