Building Data Views

Building Data Views
Parts of the View
Table Components
Headers
Axes
Panes
Cells
Marks
Mark Types
Automatic Mark
Text Mark
Bar Mark
Square Mark
Circle Mark
Shape Mark
Line Mark
Polygon Mark
Gantt Bar Mark
Pie Mark
Stacking Marks
Example – Stacking Bars
Example – Stacking Lines
Changing Mark Size and Color
Changing Mark Size
Changing Mark Color
Titles
Captions
Field Labels
Legends
Building Views Manually
Dragging Fields
The Basics
Adding More Fields
Adding Headers Using Dimensions
Adding Axes Using Measures
Rearranging the Rows and Columns
Types of Shelves
Columns and Rows Shelves
Hide Rows and Columns
Pages Shelf
Filters Shelf
Level of Detail Shelf
Color Shelf
Categorical Colors
Quantitative Colors
Transparency
Effects
Size Shelf
Categorical Sizes
Example – Categorical Sizes
Quantitative Sizes
Example – Quantitative Sizes
Shape Shelf
About Shapes
Editing Shapes
Custom Shapes
Tips for Creating Custom Shapes
Label Shelf
Path Shelf
Working with Large Views
Example – Building Data Views Manually
Building Views Automatically
Show Me!
Add to Sheet: Double-Click
Using Multiple Measures
Indivudal Axes
Blended Axes
Dual Axes
Combination Charts
Filtering
Adding Filters
Selecting Data to Filter
Selecting Headers
Selecting Marks
Dragging Fields to the Filters Shelf
Filtering Dimensions
Basic Categorical Filters
Adding Conditions to Filters
Adding Limits to Filters
Example – Filtering Dimensions
Filtering Measures
Basic Quantitative Filters
Showing and Hiding Values in the Filter Dialog Box
Example – Filtering Measures
Filtering Dates
Relative Date Filters
Range of Dates
Other types of Date Filters
Discrete Date Filters
Example – Filtering Dates
Using Quick Filters
Turning on Quick Filters
Quick Filter Options
General Quick Filter Options
Categorical Quick Filter Options
Quantitative Quick Filter Options
Customizing Quick Filters
Searching Quick Filters
Global Filters
Context Filters
Creating a Context Filter
Example – Context Filters
Calculation Filters
Sorting, Grouping, and Sets
Sorting
Computed Sorting
About Computed Sorting
How to Sort Data (Computed Sorts)
Example – Sorting a Text Table
Example – Sorting a Hierarchy
Manual Sorting
Sorting using the Toolbar
Sort by Drag and Drop
Example- Manually Sorting Drawing Order
Groups
Creating Groups
Editing an Existing Group
Finding Members in the Groups Dialog Box
Sets
About Sets
How to Create a Set
Create a Set by Selecting Marks
Create a Set from a Field
Create a Nest Set
Creating Sets Examples
Example – A Set Containing a Subset
Example – A Set Containing Unique Encodings
Example – Hierarchical Sets and their Descendents
Dates and Times
Changing Date Levels
Fiscal Dates
Perfect Pivoting with Dates
Continuous Dates
Reference Lines and Bands
Types of Reference Lines and Bands
Adding Reference Lines
Adding Reference Bands
Adding Reference Distributions (Bullet Graphs)
Basic Reference Distributions
Bullet Graphs
Editing Reference Lines and Bands
Removing Reference Lines and Bands
Inspecting Data
Select
Zoom Controls
Pan
Undo and Redo
Drop Lines
Summary Card
View Data
Describing the View

Building Data Views
This section discusses the basics of using Tableau to build views of your data. You will learn how to build views both manually and automatically using the built in tools. Additionally, you will learn how to sort, filter, group, create sets. Finally, this section includes information about working with dates and times, adding reference lines and bands, and viewing your data.
• Parts of the View
• Building Views Manually
• Building Views Automatically
• Using Multiple Measures
• Filtering
• Sorting, Grouping, and Sets
• Dates and Times
• Reference Lines and Bands
• Inspecting Data
Parts of the View
This section describes the basic components of the views you can create in Tableau. The parts of a view can be categorized as either table components, which are part of every view, or optional components, which can be turned on or off.
• Table Components
• Headers
• Axes
• Panes
• Cells
• Marks
• Titles
• Captions
• Field Labels
• Legends
• Table Components
• Data views are displayed in a table on every worksheet. A table is a collection of rows and columns, and consists of the following components: Headers, Axes, Panes, Cells, and Marks.
• In addition to these, you can optionally show Titles, Captions, Field Labels, and Legends.
•
•
Headers
Headers are created when you place a dimension on the Rows shelf or the Columns shelf. The headers show the member names of each field on the shelves. For example, in the view below the column headers show the members of an Order Date field and the row headers show the members of a Product Category field.

You can show and hide row and column headers at anytime.
To hide headers:
• Right-click the headers in the view and select Show Header.

To show headers:
• Select the field in the view whose headers you want to show and select Show Header on the field menu.

Hiding headers can be really useful when you are working with multiple measures. For example, the view below shows both the sales and profit for each region along a single axis. You can see the view looks cluttered with the Measure Names headers showing. Because Measure Names is also indicated by the mark color, you can hide the excess headers to clean up the view.

Axes
Axes are created when you place a measure on the Rows or Columns shelf. By default, the values of the measure field are displayed along a continuous axis.

You can show and hide axes at anytime.
To hide axes:
• Right-click the axis in the view and select Show Header.

To show axes:
• Select the measure in the view whose axis you want to show and select Show Header on the field menu.

Panes
Panes are created by the intersection of the rows and columns in a table. Depending on the table type, panes might be created by the intersection of an axis with headers, an axis with an axis, or headers with headers. Panes are identified by lines within the table.

Cells
Cells are the basic components of any table you can create in Tableau. For a text table, the cell is the intersection of a row and a column, and is where the text is displayed. For other view types such as bar charts and scatter plots, identifying the cell is not always possible or useful.

Marks
Tableau does not use chart types to build data views. Instead, data are displayed with marks, where every mark corresponds to a row (or a group of rows) in your data source.
You can build views of your data by placing fields on shelves and by selecting the appropriate mark type (or by accepting the default mark type).
• Mark Types
• Stacking Marks
• Changing Mark Size and Color
Mark Types
Mark types are available from the Mark menu. All mark types can be modified by color-encoding and by size-encoding (except polygon) the data

• Automatic Mark
• Text Mark
• Bar Mark
• Square Mark
• Circle Mark
• Shape Mark
• Line Mark
• Polygon Mark
• Gantt Bar Mark
• Pie Mark
• Automatic Mark
• When the Mark menu is set to Automatic, Tableau automatically selects the best mark type for your data view. This mark type is determined by the inner fields on the Rows and Columns shelves.
•
•
•
• For example, if you create a view with a dimension as the inner field on both the Rows shelf and the Columns shelf, the text mark is automatically selected. If you create a view that has measures on both the Rows shelf and theColumns shelf, the shape mark is automatically selected. If you create a view with a dimension as the inner field on the Rows shelf and a measure on the Columns shelf (or vice versa), the bar mark is automatically selected. Note that Tableau automatically places measures inside dimensions when they share a shelf.
• You can override the default selection and use any mark type that provides insight into your data. However, you should exercise some caution when manually selecting a mark type because the resulting view might hide important information about your data.
Text Mark
The text mark type is useful when you want to display the numbers associated with one or more dimension members. This type of view is often called a text table, a cross-tab, or a Pivot Table. Tableau displays your data using text when:
• The Mark menu is set to Automatic, and you place one or more dimensions as the inner fields on both theRows and the Columns shelves.
• You select Text from the Mark menu.
Initially, the data are displayed using the

icon.

To complete the view, you must place a field (typically a measure) on the Text shelf. As shown below, the Salesmeasure, which is aggregated as a summation, is used to complete the table.

Note:
You can create a cross-tab of any data view by selecting the Edit > Duplicate as Cross-tab menu item.
Because of the flexibility of Tableau, you might create a view that contains overlapping text. In this case, the following warning dialog box appears. If you do not want to display this dialog box in the future, select the check box in the lower left. To display the dialog box again, select the Help > Show Messages Again menu item.

Overlapping text occurs when multiple data source values contribute to a single text table cell. There are three common cases to consider.
• Level of detail – If you place a dimension on the Level of Detail, Color, Shape, Size, or Text shelf, overlapping text occurs if multiple dimension members (levels of detail) contribute to a text table cell. To avoid overlapping text in this case, you might consider placing the dimension on the Rows or the Columns shelf.
• Disaggregated data – If you disaggregate a measure placed on the Text shelf, overlapping text occurs if multiple data source rows contribute to a text table cell. If you want to display disaggregated data, a text table is probably not the best choice. Instead, consider displaying the data in a scatter plot.
In the example below, overlapping text occurs when you disaggregate the Sales measure. As shown below, the cells contain overlapping sales data. This is because more than one data source row has a sale record for a given year and product. Note that Office Machines in 2004 indicates that there is only one sales record. However, this cell can still contain overlapping text if there are multiple data source rows with the same value. In this case, the overlapping text warning dialog box would still appear.

Bar Mark
The bar mark type is useful when you want to compare measures across categories, or when you want to break data down into stacked bars. Tableau displays your data using bars when:
• The Mark menu is set to Automatic, and you place a dimension and a measure as the inner fields on theRows and Columns shelves (or vice versa). If the dimension is a date dimension, the Line mark is chosen instead.
• You select Bar from the Mark menu.
Note that the marks are automatically stacked.
The data view shown below displays a dimension and a measure and is color-encoded by a dimension. Because theMark menu is set to Automatic, the data are displayed using bars.

Square Mark
The square mark type is useful when you want to clearly see individual data points. When you select Square from theMark menu, Tableau displays your data using squares.
The data view shown below displays several dimensions in both the rows and columns of a table. If the Mark menu was set to Automatic, the data would be displayed using text. By manually selecting Square, a completely different view is created. In particular, by placing a measure on the Color shelf, square marks can be used to create a heat map.

To reproduce this view, select the Format > Cell Size > Square Cell menu item and then adjust the size of the squares using the Size slider.
Because of the flexibility of Tableau, you might create a view that contains overlapping data and is difficult to interpret. One way to do this is to place a dimension on the Color shelf. A view with overlapping data can be deceptive because only one of the marks for each cell is visible.
For example, suppose you replace the Profit measure in the example above with the Container dimension. As shown below, the squares indicate that there aren’t any products shipped by Small Pack (brown) or Wrap Bag (pink).

Filter Container to only include Small Pack and Wrap Bag.

As you can see, all of the squares have changed color showing that the marks overlapped.

Circle Mark
When you select Circle from the Mark menu, Tableau displays your data using circles.
As shown below, the data are displayed using circles. If the mark type was set to Automatic, Tableau would display the data using a shape (an open circle).

Shape Mark
The shape mark type is useful when you want to clearly see individual data points while also viewing categories associated with those points. Tableau displays your data using a shape when:
• The Mark menu is set to Automatic, and you place one or more measures on both the Rows and the Columnsshelves.
• You select Shape from the Mark menu.
The view shown below displays the data from two measures. Because the Mark menu is set to Automatic, the data are displayed using a shape.

By default, the shape used is an open circle. You can select a different shape by clicking on the shape legend. As shown below, twenty unique shapes are available.

To enhance the data view, you can place a dimension on the Shape shelf. Tableau separates the marks according to the members in the dimension, and assigns a unique shape to each member. The shape legend displays each member name and its associated shape.
As shown below, the Ship Mode dimension is used to shape-encode the view.

Line Mark
The line mark type is useful when you want to see trends in data over time, your data are ordered, or interpolation makes sense. Tableau displays data using lines when:
• The Mark menu is set to Automatic, and you place one or more measures on either the Columns shelf or theRows shelf, and then plot the measures against a date dimension or a continuous dimension.
• You select Line from the Mark menu.
The data view shown below displays a dimension in the column of a table and several measures as the rows of the table.

With the line mark type, you can specify the drawing order of the line by placing a field on the Path shelf.
As the density of data increases, trends are often easier to see when using lines. This view shows 90 data points.

Polygon Mark
Polygons are points connected by lines. The polygon mark type is useful when you want to connect points to create data areas. Tableau displays data using polygons when you select Polygon from the Mark menu.
Note:
The polygon mark is not commonly used and often requires a specially constructed data source.
The view shown below comes from a specially constructed data source that holds geographic and election data. It displays the 48 contiguous US states as a function of latitude and longitude and color-encodes each state by the 2000 presidential election results.
If Mark is set to Automatic, the data will be displayed using a shape. By manually selecting Polygon, and adding additional fields to the view, a different view is created.

Every state is considered to be a polygon in the data source. The PolygonID field on the Level of Detail shelf is distinct for each US state. You can remove states from the view by filtering this field.
Additionally, you can specify the drawing order of the lines that constitute each polygon by placing a field on the Pathshelf. In this example, the PointOrder measure is used to draw each state.
Gantt Bar Mark
The Gantt bar mark type is useful when you want to view dates, project plans, or the relationships between different quantitative variables. Tableau displays your data using Gantt bars when:
• The Mark menu is set to Automatic and you place one or more dimensions on either the Columns shelf or the Rows shelf, and then plot the dimensions against a continuous quantity.
• You select Gantt Bar from the Mark menu.
The distinguishing characteristic of Gantt bars is that the length of every mark is proportional to the measure placed on the Size shelf.
The data view shown below displays a dimension as a function of a continuous date. If the Mark menu is set toAutomatic, the data would be displayed using bars. By manually selecting Gantt Bar and adding additional fields to the view, a different view is created.

In particular, by placing the Time to Ship measure on the Size shelf, every bar in the view has been drawn with a length that indicates the delivery time of an order. Additionally, by placing the Ship Mode dimension on the Color shelf, each bar is color-encoded by the ship mode.
Pie Mark
The pie mark can be used to show proportions. Although generally this type of information can be better shown using stacked bar charts, there are cases where using pie marks can be very effective. For example, pie marks are very useful when trying to convey the percentage allocation of marketing expenses by state where the spending of geographically close states are very relevant.
Tableau will never use the pie mark as an automatic mark type, but you can select Pie on the Mark menu.
When you select the Pie mark type, an additional shelf is available for angle. The Angle shelf determines the angular measure of the pie wedges. For example if you place a measure such as Sales on the angle shelf, the total 360 degrees of the pie corresponds to the total sum of sales and each wedge is divided for the values of the field on the Color shelf.
The view below shows the time it took to ship products by various ship modes. The data overlays a map and shows the information by zip code. We can quickly see that Regular Air takes the longest to ship in this particular region except in the south part of Michigan where they seem to have optimized for that ship mode.

Stacking Marks
Stacking marks is relevant when your data view includes numeric axes. That is, at least one measure has been placed on the Rows or Columns shelf. When marks are stacked, they are drawn cumulatively along an axis. When marks are not stacked, they are drawn independently along an axis. That is, they are overlapping.
Stacking marks is particularly useful for bar charts which is why Tableau automatically stacks bars. You might find that stacking marks is useful for other marks such as lines as well. You can control whether marks are stacked or overlapping in any given view by selecting the Analysis > Stack Marks menu item. You can either allow Tableau to automatically select whether the marks are stacked or you can specify on or off. The default mode is automatic. When you are in automatic mode, the Stack Marks menu indicates whether stacked marks is on or off.

If you select On or Off on the Stack Marks menu, you are switched to manual mode. Your selection remains throughout any changes you make to the view.
The following examples illustrate stacking marks.
• Example – Stacking Bars
• Example – Stacking Lines
• Example – Stacking Bars
• Consider the stacked bars view shown below. It was created by placing a dimension on Columns shelf, placing a measure on the Rows shelf, and color-encoding the data by a dimension.
•
•
•
• Because the mark type is a bar, Tableau automatically stacks the marks. This means that the marks are drawn cumulatively and the height of each stacked segment within each bar represents the value for that segment. For example, the sum of the profit for products shipped by Express Air (orange bar segment) in the Corporate market is $68,450.
• If you un-stack the marks, they all start from the horizontal axis. As shown below, you can still view the individual bar segments. Be aware, however, because un-stacked marks overlap, it is possible to create a view where bar segments are not visible.
•
•
• Example – Stacking Lines
• Consider the data view shown below. It was created by placing a date dimension on the Columns shelf, placing a measure on the Rows shelf, and color-encoding the data by a dimension. Because the mark type is a line, the marks are not automatically stacked. Instead, they are drawn independently from the horizontal axis.
•
•
•
• Interpret any data point by reading the associated values from the horizontal and vertical axes. For example, in the year 2007, the Corporate (light blue) sales totaled $166,269. That is, the space between that data point and the horizontal axis is equal to the sum of the sales for the Corporate market.
• Now, stack the marks by selecting the Analysis > Stack Marks > On menu item. The stacked lines view is shown below.
•
•
•
• In this view, the lines are no longer independent of each other. Instead, they are drawn cumulatively. The stacking order is given by the order of the dimension members in the data source. This order is reflected in the color legend, from bottom to top.
• Therefore, the stacked Small Business (teal) line is the same as its un-stacked version because it’s at the bottom of the stacking list. The stacked Home Office (peach) line is derived by adding its un-stacked values to the un-stacked Small Business values. The stacked Corporate (light blue) line is derived by adding its un-stacked values to the stacked Home Office data. The stacked Consumer (blue) line is derived by adding its un-stacked values to the stacked Corporate data.
• The vertical axis gives the new scale for the stacked marks. Interpret the space between consecutive lines as the sum of the profit. The lines are no longer all compared to the horizontal axis.
• For example, notice that the tooltip for the 2007 Corporate data still shows the profit as $166,269. The interpretation is that the space between the Corporate data and the Home Office data yields the sum of the profit for the Corporate market.
Changing Mark Size and Color
You can format marks by changing the mark size and color. This allows you to highlight specific data, to distinguish between marks effectively, and to create optimal presentations. You can also display or remove mark borders. This section discusses the following topics:
• Changing Mark Size
• Changing Mark Color
• Changing Mark Size
• Each mark is displayed with a default mark size. You can change the size of marks at any time by moving the Sizeslider.
•
•
•
• If you move the slider to the right, marks get larger. If you move the slider to the left, marks get smaller. The Sizeslider affects different marks in different ways, as described in the following table.
Mark Type Description
Circle, Square, Shape, Text Moving the slider makes the mark bigger or smaller.
Bar, Gantt Bar Moving the slider makes bars wider or narrower.
Line Moving the slider makes lines thicker or thinner.
Polygon You cannot change the size of a polygon.
Pie Moving the slider makes the overall size of the pie bigger and smaller.
• The size of your data view is not modified when you change marks using the Size slider. However, if you change the view size, the mark size might change to accommodate the new formatting. For example, if you make the table bigger, the marks might become bigger as well.
• Note:
• Changing the mark size is not the same as size-encoding the data using the Size shelf.
• Changing Mark Color
• Each mark is displayed with a color, which is presented in a color legend on the Tableau interface.
• By default, all marks use the same color. However, you can display more than one color by placing a dimension or a measure on the Color shelf (Ctrl+Alt+O). Placing a dimension on the Color shelf separates the marks according to the dimension members and assigns a unique color to each member. Placing a measure on the Color shelf creates a continuous range of colors.
• Depending on your data view, Tableau will use one of the four color legends described in the following table.
Legend Type Description
This is the default color. It is used when the Color shelf is not populated with a field. To edit the default color, select Format > Marks and modify the color in the Format window.
This legend appears when the Color shelf is populated with a dimension. To edit a color, double-click anywhere in the legend.
This is a diverging color legend and appears when the Color shelf is populated with a measure that contains both positive and negative numbers. To edit the colors, click any part of the color spectrum.
This legend appears when the Color shelf is populated with a measure that contains only positive or only negative numbers. To edit the colors, click any part of the color spectrum.
Titles
You can add a title to any worksheet or dashboard. The title is displayed on the Title card.
To show and hide titles:
• Select View > Title or click View Cards

• on the toolbar and then select the Title card.

Worksheet Title

Dashboard Title
By default, the title is the name of the sheet, but you can use a custom title and even include automatic text such as page number and sheet name.
To edit titles:
1. Right-click on the title and select Edit Title.
2. In the Page Setup dialog box, type a new title into the Title text box. Use the arrow to the right of the text box to add automatic text such as page number, sheet name, page count, and more.

You can format the font, alignment, shading, and border of titles.
Captions
All views can have a caption that is either automatically generated or manually created. The caption is displayed on the Caption card. To show the caption, select it on the View Cards toolbar menu

or select View > Caption.

The caption is automatically generated by default, however, you can edit the caption by double clicking the Caption card and selecting Manual in the subsequent dialog box.

Use the arrow to the right of the text box to add automatic text such as page number, sheet name, page count, and more.
The caption is part of the Page Setup settings and can optionally be printed and published with the view. Additionally, when you export the view as an image to another application like Microsoft PowerPoint, you can select to include the caption.
You can format the font, alignment, shading, and border of captions.
Field Labels
Placing discrete fields on the rows and column shelves creates headers in the view that display the members of the field. For example, if you place a field containing products on the rows shelf, each product name is shown as row headers. In addition to showing these headers, you can show field labels, which are labels for the headers. In this example, the rows are labeled as Product Category, thus indicating that the list of products are members of the Product Category field.

Field labels apply only to discrete fieldsdimensions. When you add continuous fields to the view, an axis is created. The axis is labeled with a header.
By default, field labels are shown. You can hide or show field labels at anytime.
To show and hide field labels:
• Select Table > Field Labels for Columns or Field Labels for Rows.

You can format the fonts, alignment, shading, and separators for field labels.
Legends
When you add fields to any of the encoding shelves such as the Color, Shape, and Size shelves, a legend appears to indicate how the view is encoded with relation to your data.

Not only do legends help you understand encodings, you can also use legends to sort, filter, and highlight specific sets of data.
Building Views Manually
Building views in Tableau can be really easy if you understand some basic concepts of how it all work. This section discusses the following topics:
• Dragging Fields
• Types of Shelves
• Working with Large Views
• Example – Building Data Views Manually
Dragging Fields
You can build views of your data by dragging fields from the Data window to the view. You can drag fields to a variety of active areas in the view or place them on the shelves that are part of every worksheet.
• The Basics
• Adding More Fields
• Rearranging the Rows and Columns
• The Basics
• When you begin creating a new data view on a blank worksheet, drag a field from the Data window to and drop it in the view.
•
•
•
• While dragging fields you can pause on the active areas in the view to see how the field will be added to the view. For example, in general dimensions will add row and column headers to the view while measures add continuous axes. Below are some examples of how fields can be added to the view.
•
•
•
• For a more advanced discussion of dimensions and measures, refer to .
• When you drag a field to one of the active areas in the view, the field is added to the view and displays on one of the shelves. For example, in the view below the Regions are shown as Rows and Profit is shown as a Column with an continuous axis.
•
•
•
• You can drag fields directly to the shelves instead of the active areas in the view. You can also drag fields from one shelf to another shelf. The number of fields that you can place on the Columns, Rows, Level of Detail, Filters, andPages shelves is unlimited. However, the Color, Size, Shape, Text, and Path shelves can hold only one field at a time. Refer to for more information about each of these shelves.
• To remove a field from a shelf, drag it off the worksheet or select Remove on the field’s context menu. To quickly remove multiple fields from a shelf, right-click the shelf and select Clear Shelf on the context menu.
Adding More Fields
You can add as many fields as necessary by dragging and dropping them on the different areas of the view. Once there are more fields in the view there are some extra active areas. For example you can add replace fields by dropping them on existing headers and axes in the view. Or instead of replacing the field you can blend multiple measures onto a single axis. Finally, you can rearrange the rows and columns in the view.
• Adding Headers Using Dimensions
• Adding Axes Using Measures
• Adding Headers Using Dimensions
• You can add headers by dragging a dimension and dropping on either side of existing headers, or to the left of an axis. For example, in the view below you can add the Region field by dragging it and dropping it to the right of the product names.
•
•
•
• You can see that as you hover over the view, a dotted black line indicates active areas where you can add headers.
Adding Axes Using Measures
You can add axes by dragging a measure and dropping it on an active area in the view. If an axis already exists in the view you can replace the existing axis, blend the new measure with the existing axis, or add a secondary axis.
Replace the Existing Axis
Drag the new measure to the top left portion of the axis in the view. A small square drop zone appears and a single axis icon displays to indicate that a single axis will be left when you drop the measure.

Blend the Measures on Single Axis
You can show multiple measures on a single axis by dragging the new measure directly on top of the existing axis. Blending measures uses the Measure Names and Measure Values fields. For a more details example of blending measures refer to .

Add a Secondary Axis
Drag the field to the right side of the view to add the measure as a secondary axis. Secondary Axes are useful when you want to compare two fields that have different scales. In this case, blending the these axes would distort the view. Instead you can add a secondary axis. You can add up to four axes to the view: two on the Columns shelf and two on the Rows shelf. Below is an example of a secondary axis view showing the Dow Jones Industrial Average and NASDAQ close values over time.

Rearranging the Rows and Columns
Finally, you can rearrange the rows and columns in the view by dragging the selection border for headers or an axis.

Types of Shelves
Every worksheet in Tableau contains shelves. By placing fields on shelves, you can create the rows and columns of a data view, exclude data from the view, show additional levels of detail, and encode the data in various ways.
Each section contains examples that illustrate how a simple data view is modified by placing a dimension or a measure on the shelf.
Some shelves are available only when certain mark types are used. For example, the Shape shelf appears when the shape mark type is used. Additionally, some shelves are not particularly useful with certain mark types. Refer to for more information about marks.
You should experiment with various combinations of shelves, fields, and mark types to find the optimal view for your data. Tableau can also help you determine the best way to display your data using Show Me! Refer to to learn more.
• Columns and Rows Shelves
• Pages Shelf
• Filters Shelf
• Level of Detail Shelf
• Color Shelf
• Size Shelf
• Shape Shelf
• Label Shelf
• Path Shelf
Columns and Rows Shelves
The Columns shelf creates the columns of a table, while the Rows shelf creates the rows of a table. You can place an unlimited number of fields on these shelves.
When you place a dimension on the Rows or Columns shelf, headers for the members of that dimension are created. When you place a measure on the Rows or Columns shelf, quantitative axes for that measure are created. As you build up your data view with more fields, additional headers and axes are included in the table and you get an increasingly detailed picture of your data.
In the view shown below, the members of the Customer Segment dimension are displayed as column headers, while the Profit measure is displayed as a vertical quantitative axis.

Tableau displays data using marks, where every mark corresponds to a row (or a group of rows) in your data source. The inner fields on the Rows and Columns shelves determine the default mark type. For example, if the inner fields are a measure and a dimension, the default mark type is a bar. You can manually select a different mark type using theMark menu. Refer to id845ffa42-31fa-49bf-8ca9-477816c899de.html#i1000296 for more information.
Adding more fields to the Rows and Columns shelves adds more rows, columns, and panes to the table.

• Hide Rows and Columns
Hide Rows and Columns
Generally you will add dimensions and measures to create the rows and columns of the table and you’ll either include all data or add filters to only show a subset. However, when you filter data it is also excluded from calculations and other computions performed on the summarized data in the table. For example, depend on the data shown in the view for computations such as year over year growth and running totals. In these cases you can hide the rows and columns that you don’t want to show without changing the calculation.
To hide a row or column:
• Right-click the row or column you want to hide and then select Hide.

To show hidden data:
• Open the field menu for a field that has hidden columns or rows and select Show Hidden Data.

Hiding columns is especially useful when using table calculations that compare to previous or next. In that case, there is always a row or column that doesn’t show data because there is no data to compare to. You can simply hide the empty column without modifying the table calculation.
Pages Shelf
The Pages shelf lets you break a view into a series of pages so you can better analyze how a specific field affects the rest of the data in a view. When you place a dimension on the Page shelf you are basically adding a new row for each member in the dimension. When you place a measure on the Pages shelf, the measure is converted into a discrete measure.
The page shelf creates a view on a different page for each new row so you can easily flip through each view and compare them on a common axis. For example, the view below shows the Profit vs. Sales by Region for each day throughout the month.

You can see that it is difficult to see how these two measures have interacted from day to day. However, when you move the Day field to the Pages shelf and flip through the pages (one for each day) you can quickly discover hidden insights. In this example, it is interesting that the 19th is an especially big day in terms of sales and profit in the Western region.

When you add a field to the page shelf the Current Page card displays. Use this card to navigate through the pages.

There are three ways to navigate through the pages in a view.
Jump to a specific page
Select the member or value you want to view from the drop-down list on the Current Page card to display a specific page rather then scrolling through the entire sequence.

Manually Advance through the pages
You can manually advance through the sequence of pages by doing one of the following:
• Use the forward and back buttons on either side of the drop-down list to navigate through the pages one at a time.
• Use the Page Slider to quickly scroll forward and backward in the sequence of pages.
• Use the keyboard shortcuts below to scroll forward and backward in the sequence of pages.
F4 Starts and stops forward playback
SHIFT + F4 Starts and stops backward playback
CTRL + . Skip forward one page
CTRL + , Skip backward one page
Automatically Advance through the pages
Use the playback controls to watch a slide show of the pages in the view. You can play forward, play backward, and stop. You can control the speed of playback with the speed controls in the bottom right corner of the card. The smallest bar indicates the slowest playback speed.

Page History
Optionally show page history using the Show History checkbox. When you show history, marks from previous pages are shown in addition to the previous page. Open the drop-down control for history to specify what marks to show and when to show them.

The history drop-down control has the following options:
• Marks to show history for – select whether you want to show history for just selected marks, highlighted marks, marks that you’ve manually selected to show history for, or all marks. You manually show history for marks by right-clicking the mark in the view and selecting and option on the Page History menu.
• Length – select the number of pages to show in the history.
• Show – specify whether to show the historical marks, a line tracing through the previous values (trails), or both.
• Marks – format the historical marks including the color and how much to fade them If the color is set to automatic, the marks will either use the default mark color or the color encoding on the Color shelf.
• Trails – format the lines that are drawn through the historical marks. This options is only available if Trails is selected in the Show options.
Note:
Page trails may not display if there are multiple marks per color on a page. Make sure that the level of detail for the view is less than or equal to the level of detail on the pages and color shelves.
Filters Shelf
The Filters shelf allows you to specify which data to include and exclude. For example, you might want to analyze the profit for each customer segment, but only for certain shipping containers and delivery times. By placing fields on theFilters shelf, you can create such a view.
Note:
This section presents a brief overview of filtering. Refer to for a complete description.
You can filter data using measures, dimensions, or both at the same time. Additionally, you can filter data based on the fields that make up the columns and rows of the table. This is called an internal filter. You can also filter data using fields that don’t contribute headers or axes to the table. This is called an external filter. All filtered fields display on theFilters shelf.
To illustrate the basic concepts of filtering, consider the following view.

Suppose you are not interested in the Small Business data. You can remove this column from the view by filtering theCustomer Segment dimension. To do so, select Filter from the field’s context menu. The Filter dialog box opens. By default all members are selected. Un-check Small Business to exclude it from the view. All selected members will be included.

As shown below, Customer Segment is automatically placed on the Filters shelf, and the view now contains three columns instead of the previous four.

Suppose you want to only view sales for products that were shipped in boxes. To do this, place the Containerdimension directly on the Filters shelf. This is an example of an external filter because Container is not part of the view. That is, it does not contribute row or column headers.
The Filter dialog box shown below automatically opens. By default, none of the members are selected. Select the members you want to keep as part of the view. All deselected members are excluded.

The modified data view is shown below. The tooltip shows that the sum of the sales for the Consumer segment has decreased to $2,225,449. This number is derived by summing all the rows in the data source that are associated with the Corporate market and that use a box as a shipping container.

The order of fields placed on the Filters shelf does not affect the data view because the filters are independent. The result of filtering by customer segment, and then by container is the same as filtering by container and then by customer segment.
Level of Detail Shelf
Whenever you place a dimension on the Rows or Columns shelf, the categorical members create table headers. The headers represent levels of detail because they separate the data source rows into specific categories. You can identify each category by the member name. For example, the Customer Segment dimension separates the data source rows into four levels of detail: Consumer, Corporate, Home Office, and Small Business.
The Level of Detail shelf also allows you to separate the marks in a data view according to the members (levels of detail) of a dimension. However, unlike using the Rows and Columns shelf, using this shelf is a way to show more data without changing the table structure.
As shown below, the bars are separated into segments according to the members of the Product 2 – Sub-Categorydimension. The size of each segment reflects the contribution to the profit for a particular member. For example, the view below shows that Appliances category in the Corporate market has a profit of $50,960.

You can place any number of dimensions on the Level of Detail shelf. In fact, placing all dimensions on this shelf is one way to display all the rows of your data source.
Note:
The Level of Detail shelf works only if the measures that contribute axes to the table are aggregated. If the measures are disaggregated, then it isn’t possible to separate the marks into additional levels of detail because all levels of detail are already shown.
Also, placing a measure on the Level of Detail shelf has no effect because measures do not contain members. However, you can place measures on this shelf if you want to export their values to Microsoft Access, copy their values to the Windows Clipboard, or view them in the tooltips.
Color Shelf
All marks have a default color that is used when there are no fields on the color shelf. Most marks use a blue color while text marks are shown in black.
The Color shelf encodes data by assigning different colors to the marks in a data view based on the values of a field. The effect of color-encoding your data view depends on whether you use categorical or quantitative colors. You can also use the drop-down control to specify other color properties such as transparency, borders, and halos. The color shelf is discussed in the following topics:
Note:
Color encodings are shared across multiple worksheets that use the same data source to help you create consistent displays of your data. For example, if you define the Western region to be green, it will automatically be green in all other views in the workbook. You can set the default color encodings for a field by right-clicking the field in the Data window and selecting Edit encodings > Color.
• Categorical Colors
• Quantitative Colors
• Transparency
• Effects
• Categorical Colors
• When you add a dimension to the Color shelf a categorical legend is added based on the members in the dimension field. You can modify the colors used in the legend by right-clicking on the legend and selecting Edit colors or by double-clicking on the legend. The Edit Colors dialog box for a categorical legend is shown below.
•
•
•
• To change the color of a member, select the member on the left and then select the new color in the palette on the right. When finished, click OK to close the format dialog box.
• You can select a different color palette from the drop down list in the upper right of the Edit Color dialog box. Select from either categorical palettes or ordinal palettes. A categorical palette, such as Tableau 20 contains several distinct colors that can be assigned to dimension members that have no inherent order. Ordinal palettes contain a spectrum of related colors, which can be used for dimension members that have an associated order such as dates and numbers. The views below show a categorical palette versus an ordinal palette.
• Once you select a palette, click Assign Palette to automatically assign the new palette colors to the members in the field. When finished, click OK to view the changes and close the dialog.
• To return to the automatic color settings that Tableau provides by default click Reset in the Edit Colors dialog box and then click OK.
Quantitative Colors
When you add a measure to the Color shelf a quantitative legend is added creating a continuous range of colors. You can modify the colors used in the range, the distribution of color, and other range attributes in the Edit Color dialog box. Right-click the legend and select Edit Colors or double-click on the legend. The Edit Colors dialog box for a quantitative legend is shown below.

To change the color used in the range, simply click on the color indicator to the right of the range and click on a new color in the spectrum. You can select a new palette from the Palette menu. You can choose between a sequential palette and a diverging palette. A sequential palette shows a simple range of values using color intensity to indicate one end of the range from the other. A diverging palette shows two ranges of values using color intensity to show the magnitude of the number and the actual color to show which range the number is from. Diverging palettes are most commonly used to show the difference between positive and negative numbers. When finished, click Apply.
Each of the options for formatting quantitative colors are described below:
Using Stepped Color
You can modify how the colors are distributed by selecting Stepped Color. The stepped color option groups the values into uniform bins each given a unique color. Use the text box to specify how many bins you want to use. For example, if you had a range of values from 0 to 100 and you select 5 steps, the color range would be broken up every 20 units. That means that all points between 0 and 20 would be colored the same, all points between 21 and 40 would be colored the same and so on. The dialog box below shows the color range broken up into five steps. When finished, click Apply.

If a diverging color palette is selected, the center point is shown on the color ramp with a small black mark. When the number of steps is odd, the center mark is placed in the middle of the center step. When the number of steps is even, the center mark is placed at the boundary of the center-most two steps.
Reversing the Color Palette
Select Reversed to switch the order of colors in the range. For example, if you want lower values to have a darker intensity in a sequential palette, reverse the palette. Alternatively, if you are using a diverging color palette with red representing -100 to 0 and blue representing 0 to 100, you can switch the colors using the reverse option to make blue represent the negative range and red represent the positive range. When finished, click Apply.
Using the Full Color Range
When you are using a diverging color palette you can select to Use Full Color Range. When you select this option, Tableau assigns the starting number a full intensity and the ending number a full intensity. If the range is from -10 to 100, the color representing negative numbers changes in shade much more quickly than the color representing positive numbers. If you do not select Use Full Color Range, Tableau assigns the color intensity as if the range was from -100 to 100 so that the change in shade is the same on both sides of zero. The example below shows a diverging color palette for values from -10 to 150. Without using the full color range, -10 is represented by a light red color. When the full color range is used, -10 is represented by a full red. When finished, click Apply.
Limiting the Color Range
You can limit the range that the colors are distributed across using the Advanced options. When you click Advanced in the Edit Colors dialog box, you can select to specify the start, end, and center values on the range by selecting the check box and typing a new value into the textbox. The Start value is the lower limit in the range, the End value is the upper limit, and the Center value is the where the neutral color is located on a diverging color palette. When finished, click Apply.
Resetting the Color Range
To return to the automatic color settings that Tableau provides by default click Reset in the Edit Colors dialog box and then click OK.
Transparency
You can also modify the transparency of the marks drop-down control next to the Color shelf. This is especially useful in dense scatter plots or when you are looking at data overlaying a map or background image. As you slide the slider toward the left the marks become more transparent.
Effects
Use the drop-down control next to the color shelf to modify other color properties. You can
Mark Borders
By default, Tableau displays all marks without a border. You can turn on the mark borders for all mark types except text, line, and shape. Turn on mark borders by selecting a color on the color shelf drop-down control.
Borders are often useful for distinguishing between closely spaced marks. For example, the view shown below has mark borders turned on (left) and turned off (right). As you can see, when borders are turned off, the marks become indistinguishable in the areas where they are tightly clustered.

Note:
You can also use transparency to show the density of marks.
Leaving mark borders off is particularly useful when you are viewing a large number of small marks that are color-encoded. It can be difficult to see the color encoding because the borders dominate the marks.
For example, the view shown below displays bars that are segmented by a large number of color-encoded dimension members. As you can see, when mark borders are turned on some marks are difficult to identify by color. When borders are turned off, the marks can easily be distinguished.

Mark Halos
In order to make the marks in a view more visible when placed on top of a background image or map, each mark is surrounded by a solid contrasting color called a halo. Mark halos are available when you have a background image or background map. You can turn mark halos by selecting a color on the color shelf drop-down control.
The view below uses a map so the marks are surrounded by orange halos to make them stand out.

Markers
When you are using the Line mark type, you can add a marker effect to show and hide the points along the line. You can show selected points, all points, or no points. Select a marker effect on the color shelf drop-down control.
Size Shelf
The Size shelf allows you to encode data by assigning different sizes to the marks in a data view. Depending on whether you use a discrete or continuous field you will add either categorical or quantitative size encodings. This section discusses the following topics:
• Categorical Sizes
• Example – Categorical Sizes
• Quantitative Sizes
• Example – Quantitative Sizes
Categorical Sizes
When you place a discrete field on the Size shelf, Tableau separates the marks according to the members in the dimension, and assigns a unique size to each member. Because size has an inherent order to it (small to big), categorical sizes work best for ordered data like years or quarters.
Note that size-encoding data with a discrete field separates the marks in the same way as the Level of Detail shelf does, and then provides additional information (a shape) for each mark. When you add categorical size encoding to the view, a legend displays showing the sizes assigned to each member in the field placed on the size shelf. You can modify how these sizes are distributed in the Edit Sizes dialog box.
To edit categorical size encodings in a view:
1. Double-click on the legend or select Edit Size from the legend’s menu to open the Edit Sizes dialog box.

2. In the Edit Sizes dialog box, the sizes are displayed on the left and a size range slider is shown on the right. The sizes assigned to each member are distributed across the specified range. Use the slider to adjust the sizes assigned to each member.
You can also select Reversed to assign the largest mark to the smallest value and the smallest mark to the largest value.
3. When finished click OK.
4. Example – Categorical Sizes
5. The view below shows the sales and profit of a superstore broken down by region and order date. The order priority is indicated by the size of the mark.
6.
7.
8.
9. In this case, the highest priority orders are shown with the smallest mark, which doesn’t make sense. Use the Edit Sizes dialog box to Reverse the range so that the highest priority orders have the largest mark.
10.
11.

Quantitative Sizes
When you place a continuous field on the Size shelf, Tableau draws each mark with a different size using a continuous range. The smallest value is assigned the smallest sized mark and similarly the largest value is represented by the largest mark.
When you add quantitative size encoding to the view, a legend displays showing the range of values over which sizes are assigned. You can modify how these sizes are distributed in the Edit Sizes dialog box.
To edit quantitative size encodings:
1. Double-click on the size legend or select Edit Size from the legend’s menu to open the Edit Sizes dialog box.

2. In the Edit Sizes dialog box, select one of the following ways to map the sizes:
o Automatically – selects the mapping that best fits your data. If the data is numeric and does not cross zero (e.g. all positive or all negative), the ‘From zero’ mapping is used. Otherwise, the ‘By range’ mapping is used.
o By range – Uses the minimum and maximum values in the data to determine the distribution of sizes. For example, if a field has values from 14 to 25, the sizes will be distributed across this range.
o From zero – Sizes are interpolated from zero making the maximum mark size assigned to the absolute value of the data value that is farthest from zero.
3. Use the range slider to adjust the distribution of sizes. When the From zero mapping is selected, the lower slider is disabled because it is always set to zero.
4. You can optionally select Reversed to assign the largest mark to the smallest value and the smallest mark to the largest value. This option is not available if you have selected to map the sizes from zero because the smallest mark is always assigned to zero.
5. Finally, you can select the Start and End checkboxes and manually type in a beginning and end value for the range of values to modify the distribution of sizes.
6. When finished, click OK.
7. Example – Quantitative Sizes
8. The view below analyzes the time it takes to ship products based on their ship mode, order date, and the size of the order. The size of each mark represents the order quantity while the color represents the Ship Mode. Looking at the view you can quickly see that most products ship within 1 and 2 days. However, larger orders shipped via Regular Air tend to take longer, especially during the second quarter. Curiously, there are a couple of smaller orders that were shipped via Express Air that took a long time to ship.
9.
10.
11.
12. You can also change the size of the marks using the Size slider.
13.
14.
15.
16. For the line and bar mark types, the size slider controls the width of the mark. For the Gantt bar mark type, the size slider controls the length of the bar. For the other supported mark types, the size slider controls the area of the mark.
Shape Shelf
The Shape shelf allows you to encode data by assigning different shapes to the marks in a data view.This section discusses the following topics:
• About Shapes
• Editing Shapes
• Custom Shapes
• About Shapes
• When you place a dimension on the Shape shelf, Tableau separates the marks according to the members in the dimension, and assigns a unique shape to each member. The shape legend displays each member name and its associated shape. When you place a measure on the Shape shelf the measure is converted to a discrete measure.
• Note that shape-encoding data separates the marks in the same way as the Level of Detail shelf does, and then provides additional information (a shape) for each mark. The Shape shelf is available when you select the shape mark type from the Mark menu. It is the default mark type when measures are the inner fields for both the Rows shelf and the Columns shelf.
• As shown below, the marks are separated into different shapes according to the members of the Customer Segmentdimension. Each shape reflects the customer segment’s contribution to the profit and sales.
•
•
Editing Shapes
By default, ten unique shapes are used to encode dimensions. If you have more than 10 members, the shapes repeat. In addition to the default palette, you can choose from a variety of shape palettes such as filled shapes, arrows, and even weather symbols.
To edit shapes:
1. Double-click the Shape Legend or select Edit Shapes on the legend’s card menu. If there is no shape encoding, you can open the Edit Shapes dialog box by clicking the shape shelf itself and then selecting More Shapes.

2. In the Edit Shape dialog box, select a member on the left and then select the new shape in the palette on the right. You can also click the Assign Palette button to quickly assign the shapes to the members of the field.

Select a different shape palette using the drop-down list in the upper right of the Edit Shape dialog box.
Note:
Shape encodings are shared across multiple worksheets that use the same data source to help you create consistent displays of your data. For example, if you define Furniture products to be represented by a square, they will automatically be squares in all other views in the workbook. You can set the default shape encodings for a field by right-clicking the field in the Data window and selecting Edit encodings > Shape.
Custom Shapes
You can add custom shapes by adding the shape image files to the Shapes folder in your Tableau Repository located in your Documents folder. When you use custom shapes, they are saved with the workbook. That way the workbook can be shared with others.
To create custom shapes:
1. Create your shape image files. Each shape should be saved as its own file and can be in many image formats including bitmap (.bmp), portable network graphic (.png), JPEG, graphics interchange format (.gif), and so on. Refer to idd8dc0ebe-6e0c-4dc7-9184-1598358d99d7.html#i1114849 for some tips on making useful shapes.
2. Place the shapes into the My Tableau Repository folder located in your Documents folder. The shapes should be put into a new folder inside the Shapes folder. The name of the folder will be used as the name of the palette in Tableau. In the example below, two new palettes are created: Maps and My Custom Shapes.

3. In Tableau, open the Edit Shape dialog box.

4. Choose the new custom palette in the drop-down list in the upper right of the dialog box. If you modified the shapes while Tableau was running, you may need to click the Reload Shapes button so the new shapes are available in the dialog box.

5. You can either assign members shapes one at a time, or click the Assign Palette button to automatically assign the shapes to the members.

You can return to the default palette at anytime by clicking the Reset button. If you open a workbook that uses custom shapes that you don’t have, the workbook will show the custom shapes. However, you can click the Reload Shapesbutton in the Edit Shapes dialog box to use the ones in your repository instead.
Below are some examples of views that use both the default and custom shape palettes.

• Tips for Creating Custom Shapes
Tips for Creating Custom Shapes
When you create custom shapes there are a few things that you can do to improve how your shapes look and function in the view. Below are some tips to help you make good custom shapes. If you are creating your own shapes, we recommend following general guidelines for making icons or clip art.
• Suggested Size – unless you plan on using the Size shelf to make the shapes really large, you should try to make your original shape size close to 32 pixels by 32 pixels. However, the original size is dependent on the range of sizes you want available in Tableau. You can resize the shapes in Tableau using the Size shelf as well as the cell size options on the Format menu.
• Adding Color Encoding – if you plan to also use the Color shelf to encode the shapes with color, you should use a transparent background. Otherwise, the entire square of the image will be colored rather than just the symbol. GIF and PNG file formats both support transparency. GIF files support transparency for a single color that is 100% transparent, while .png supports alpha channels with a range of transparency levels available on every pixel in the image. When Tableau color encodes the symbol, the amount of transparency for each pixel will not be modified, so you can maintain smooth edges.
• File Formats – Tableau does not support symbols that are in the Enhanced Meta File format (.emf). The shape image files can be in one of the following formats: .png, .gif, .jpg, .bmp, and .tiff.
Label Shelf
The Label shelf allows you to view the numbers associated with a data view, and to encode data by assigning text labels to the marks. The effect of text-encoding your data view depends on whether you use a dimension or a measure.
• Dimension – When you place a dimension on the Text shelf, Tableau separates the marks according to the members in the dimension. The text labels are given by the dimension member names.
• Measure – When you place a measure on the Text shelf, the text labels are given by the measure values. The measure can be either aggregated or disaggregated. However, disaggregating the measure is generally not useful because it often results in overlapping text.
Text is the default mark type when dimensions are the inner fields for both the Rows shelf and the Columns shelf. Refer to for more information.
The most common view using the Text shelf is a text table, which is also referred to as cross-tab or a PivotTable.

Note:
You can display text labels with other mark types by selecting clicking Show Mark Labels on the toolbar. Refer Mark Labelsto to learn more about showing and hiding mark labels.
If you place a dimension on the Text shelf, the marks are separated and labeled according to the dimension member names. If you place a measure on the Text shelf, the marks are labelled by the values contained by the measure.
As shown below, the heights of the bars are given by the Sales measure and the labels are given by the sum of the Profit measure.

Path Shelf
The Path shelf allows you to encode data by connecting marks using a particular drawing order. You can path-encode your data using either a dimension or a measure.
• Dimension – When you place a dimension on the Path shelf, Tableau connects the marks according to the members in the dimension. If the dimension is a date, the drawing order is given by the date order. If the dimension holds words such as customer names or product types, the drawing order is given by the order of the members in the data source. You can change the order by which data points are connected by changing the sort order of the members. Refer to Sorting.
• Measure – When you place a measure on the Path shelf, Tableau connects the marks according to the values of the measure. The measure can be aggregated or disaggregated.
The Path shelf is available only when you select the line or polygon mark type from the Mark menu. Refer to Mark Types for more information.
To create a useful path-encoded view, your data table should contain at least one measure. This is because you cannot create a path that connects only categorical data (dimensions).
The view below was created using storm data from the Atlantic basin in 2005. The view uses line marks with the path determined by the date of the storm. In this example, it lets you see the path of the storm.

By placing the continuous date on the Path shelf, the lines are drawn in chronological order.
Working with Large Views
Placing dimensions with a large number of dimensions on a shelf may take a long time and generally won’t be very useful when they are added. Tableau will present you with the following dialog box with the options to make it more manageable.

If you are building a data view that involves a large amount of data, it is generally more efficient to follow this procedure:
1. Turn off automatic updates by clicking the Pause Automatic Updates button on the toolbar.
2. Place all desired fields on shelves.
3. Specify filters to restrict the data to the members of interest (refer to ).
4. Turn on automatic updates by clicking the Resume Automatic Updates button on the toolbar.
At any time a query is taking too long, you can cancel the query by clicking Cancel in the progress dialog box. For more information about canceling queries refer to .
Example – Building Data Views Manually
In this example, you will use the Sample – Superstore Sales (Excel) data source to create a view that contains two aggregated measures displayed as a scatter plot. The data are color-encoded and shape-encoded, and an additional level of detail is included. The data are also filtered.
To create the view, do the following:
1. Place the Sales measure on the Columns shelf and the Profit measure on the Rows shelf.
The measures are automatically aggregated and result in one data point. The data point is displayed using the shape mark type. Note that you are not displaying any levels of detail because dimension members are not included in the view.

2. Place the Customer Segment dimension on the Level of Detail shelf.
The original mark is now separated into four marks, where each new mark is associated with a member (level of detail) of the Customer Segment dimension.

3. Place the Region dimension on the Color shelf.
Each of the four marks are now separated into four new marks for a total of sixteen. Each new mark is associated with a member of the Region dimension, and is encoded with a unique color. The color legend displays each member name and its associated color.

4. Place the Product 1 -Category dimension on the Shape shelf and filter the dimension to exclude Technology products.
Each of the twelve marks are now separated into three new marks for a total of 48. Each new mark is associated with a member of the Product 1 – Category dimension, and is encoded with a unique shape. The shape legend displays each member name and its associated shape.
To filter the data, select Filter on the Product 1 – Category context menu. The Filter dialog box opens. DeselectTechnology to exclude it from the view.

The final view is shown below.

Building Views Automatically
Tableau contains a suite of tools designed to help you quickly create useful views. Two situations in which you would want to create views automatically are when you want:
• Better Insight – People often have difficulty mapping data to views that address their analytical or presentation needs. Tableau contains built-in rules that are used to examine data and suggest ways of looking at it. In this way the software acts as a tour guide for producing useful views of data.
• Time Savings – Building data views manually can sometimes be time consuming. Using Tableau’s automatic features can help you work faster by giving you a starting view that you can further refine manually.
The specific methods for automatically generating views of data fall into two categories:
• Show Me!
• Add to Sheet: Double-Click
Show Me!
Show Me! creates a view based on one or more selected fields. When you use Show Me! simply select fields you want to analyze in the Data window and press the Show Me! button on the toolbar. Tableau automatically evaluates the selected fields and gives you the option of several types of views that would be appropriate for those fields.
1. Select Input.
Select fields in the Data window that you want to analyze. Use the Ctrl key to make multiple selections.

2. Click Show Me!

3. on the toolbar.
4. Select the type of view you want to create in the dialog box.

Any alternative that is not grey will generate a view of your data. Tooltips describe the minimum requirements for each alternative.
5. View the Result. Tableau automatically creates a view of the data.

How does Tableau know what view to create? It examines information about the fields you selected in the Data window and makes a suggestion based on best practices for presenting data. For instance, in the example shown above, a date field was selected as well as a continuous measure. Usually the best way to view a continuous measure over time is with a line.
Add to Sheet: Double-Click
Tableau supports an additional method for automatically generating views of data called Automatic Double-Click. To use this method, double-click fields in the Data window you are interested in. Tableau automatically adds each field to the view. That is, each double-click results in an additional field added to a shelf in an intelligent way. Like Show Me!, this function leverages Tableau’s ability to make an intelligent “best guess” of how the data should be displayed.
Here’s how it works:
1. Double-clicking the Sales measure in the Data window automatically adds that field to the view in an intelligent way.

2. Double-clicking the Product 1 – Category dimension in the Data window automatically adds that field to the view in an intelligent fashion.

3. Double-clicking the Order Date dimension in the Data window automatically adds that field to the view in an intelligent way. As you double-click fields they are successively added to the view. The order in which you click fields determines the type of view created.

The following table describes some of the rules used in creating automatic views via the Double Click method.
Text Table Adding a dimension first produces a text table (or cross-tab). All subsequent clicks on fields result in refinement of the text table.
Bars Adding a measure first and then a dimension produces a bar view. All subsequent clicks result in refinement of the bar view, unless a date dimension is added, at which time the view is changed to a line.
Line Adding a measure and then a date dimension produces a line view. All subsequent clicks result in refinement of the line view.
Continuous Line Adding a continuous dimension and then a measure produces a continuous line view. Subsequent dimensions result in refinement of the continuous line view. Subsequent measures add quantitative axes to the view.
Scatter Adding a measure and then another measure produces a scatter view. Subsequent dimensions result in refinement to the scatter view. Subsequent measures will create a scatter matrix.
Maps Adding a geographic field produces a map view with latitude and longitude as axes and the geographic field on the Level of Detail shelf. Subsequent dimensions add rows to the view while subsequent measures further refine the map by adding size and color encoding.
Using Multiple Measures
There are lots of different ways to compare multiple measures in a single view. For example, you can create individual axes for each measure or you can blend the two measures to share an axis and finally, you can add dual axes where there are two independent axes layered in the same pane. In any of these cases you can customize the marks for each axis to use multiple mark types and add different levels of detail. Views that have customized marks are called combination charts.
• Indivudal Axes
• Blended Axes
• Dual Axes
• Combination Charts
• Indivudal Axes
• .Add indivdual axes for each measure by dragging measures to the Rows and Columns shelves. Each measure on the Rows shelf adds an additional axis to the rows of the table. Each measure on the Columns shelf adds an additional axis to the columns of the table. For example, the view below shows quarterly sales and profit. The Sales and Profit axes are indvidual rows in the table and have independent scales.
•
•

Blended Axes
Measures can share a single axis so that all the marks are shown in a single pane. Instead of adding rows and columns to the view, when you blend measures there is a single row or column and all of the values for each measure is shown along one continuous axis. For exmaple, the view below shows quarterly sales and profit on a shared axis.

To blend multiple measures, simply drag one measure or axis and drop it onto an existing axis.

Blending measures uses the Measure Names and Measure Values fields, which are generated fields that contain all of the measure names in your data source and all of the measure values. The shared axis is created using the Measure Values field. The Measure Names field is added to the Color shelf so that a line is drawn for each measure. Finally, the Measure Names field is filtered to only include the measures you want to blend.
Note:
Blending axes is most appropriate when comparing meausres that have a similar scale and units. If the scales of the two measures are drastically different, the trends may be distorted.
Dual Axes
You can compare multiple measures using dual axes, which are two independent axes that are layered on top of each other. Dual axes are useful when you have two measures that have different scales.For example, the view below shows Dow Jones and NASDAQ close values over time. The two axes are independent scales but the marks are layered in the same pane.

To add the measure as dual axis drag the field to the right side of the view and drop it when you see a black dashed line. You can also select Dual Axis on the field menu for the measure.

You can add up to four layered axes: two on the Columns shelf and two on the Rows shelf.
Note:
When you are using dual axes you should make sure that the two axes scales align with each other so you can make a correct comparison. You can easily line the two axes up by right-clicking the secondary axis and selectingSynchronize Axis.
Combination Charts
When working with multiple measures in a view, you can customize the mark type for each distinct measure. For example, you can create a view with a line showing a target amount across several months and a bar chart showing the actual attainment for the months. These measures can be displayed as individual axes, blended axes, or dual axes.
Because each measure can have customized marks, you can customzie the level of detail, size, shape, and color encoding for each measure too.

To customize the marks for a measure:
1. Right-click the axis for the measure you want to customize and select Mark Type and then select a custom mark type.

2. The Marks card switches into an advanced mode that shows the mark properties for the measure you customized. Any changes to the mark type, shape, size, color, and level of detail will be applied to the selected measure. For example, in the view below the marks card is showing the properties for the SUM(Sales) measure. When Region is placed on the Color shelf, the encoding and level of detail is only applied to the SUM(Sales) layer. The SUM(Sales Plan) is now broken down by Region.

When the Marks card is in the Advanced mode you can switch between each of the measures in the view using control at the top of the Marks card. Select ALL to modify properties for all measures at once.

Select Single Mark Type on the Marks card menu to make all measures use the properties that are currently showing in the marks card.

Filtering
Narrow the data shown in a view using Filter. Filters are defined by selecting specific dimension members or a range of measure values. For example a view showing product sales in four different regions may be filtered to only show three regions. In this case, the filter is created by selecting the specific regions to show. Another example may be to filter the same view to only show sales between $5000 and $20,000. For this filter a range of values is defined. In addition to these basic filters you can create complex computed filters to show the top 10 products based on sales, all products sold in the last 30 days, and so on.
All fields that are filtered display on the Filters shelf so you can quickly determine the data that has been removed from the view. Filters are applied to the view in the order they appear on the Filters shelf. However, by default filters are evaluated independently from each other, which means each filter is evaluated against the entire underlying data source and other filters are not taken into account. When working with independent filters, the order on the filters shelf does not change the results.
• Adding Filters
• Global Filters
• Context Filters
• Calculation Filters
Adding Filters
All fields that are filtered show on the Filters shelf. You can add a filter either by selecting data in the view, dragging a field to the Filters shelf, or turning on quick filters. Learn about each of these ways to add filters in the following topics:
• Selecting Data to Filter
• Dragging Fields to the Filters Shelf
• Using Quick Filters
Selecting Data to Filter
You can filter data by selecting headers or marks in the view and then selecting Keep Only or Exclude on the right-click context menu. The dimension members are removed from the view and the filtered fields are added to the Filters shelf.

• Selecting Headers
• Selecting Marks
• Selecting Headers
• When you select a table header that is part of a hierarchy, all of the next level headers are also selected automatically. For example, the view shown below consists of two unrelated dimensions placed on the Columns shelf, and two levels of the same hierarchy placed on the Rows shelf.
• The selected row headers include the East member of the Region dimension, and the Texas and Louisiana members of the State dimension. Note that when East is selected, all members from the next (inner) level in the hierarchy are automatically selected.
• The selected column headers include the Colas and Root Beer members of the Gen2, Product dimension. Note that when these outer dimensions are selected, the inner dimension members from Pkg Type are not automatically selected. This is because the Gen2, Product and PkgType dimensions are unrelated.
•
•
•
• Selecting Keep Only keeps all selected headers as shown below. The Product field is filtered to show Colas and Root Beer and the Market field is filtered to show the Eastern region as well as Texas and Louisiana in the Southern region.
•
•
•
• Selecting Exclude excludes all selected headers as shown below. The Product field is filtered to show Cream Soda, Fruit Soda, and Diet Drinks. The Market field is filtered to show the Western and Central regions along with the remaining states in the Southern region.
•
•
• Selecting Marks
• Instead of selecting headers to filter, you can filter individual marks in a view. This method is useful when you are looking at a scatter plot and you want to focus on a set of outliers or remove them so you can better focus on the rest of the data. Select individual marks or click and drag to select several marks. Then right-click and select Keep Only orExclude.
• Note:
• These options are not available if a Wildcard Match filter is already specified for the field. Refer to Filtering Dimensionsto learn more about Wildcard Match filters.
•
•
Dragging Fields to the Filters Shelf
Another way to create a filter is to drag a field directly to the filters shelf. When you add a field to the filters shelf, the Filter dialog box opens so you can define the filter. The Filter dialog box differs depending on whether you are filtering a dimension, measure, or date field.
• Filtering Dimensions
• Filtering Measures
• Filtering Dates
Filtering Dimensions
Dimensions contain discrete categorical data so filtering this type of field generally involves selecting the values to include or exclude. You can create a basic categorical filter or you can define conditions and limits to create a more complex filter definition.
• Basic Categorical Filters
• Adding Conditions to Filters
• Adding Limits to Filters
• Example – Filtering Dimensions
Basic Categorical Filters
1. Drag a field from the Data window to the Filters shelf. You can also right-click a field on any shelf and selectFilter.

2. Use the General Tab of the Filter dialog box to select the values you want to include or exclude.

Each option on the General tab is described below:
• Show More/Fewer – The contents of the Filter dialog box is affected by the filters that are already set in the view. For example, if you create a filter on the Market category that only includes the western region, when you open the States filter dialog box, you typically only want to see the relevant values (the western states). If you do want to see all the values in the data source including the ones that don’t pass the current filters, you can click the Show More toggle. The Show More/Fewer option includes and excludes data from displaying in the Filter dialog box so you can find what you are looking for quicker.
• Values Selector – Use the Values drop-down list to choose a method of selecting values. Depending on the data source you are using and the type of dimension you are filtering, you can select from the following options:
o Select from List – select from a list of the values (requires a database query to get the values)
o Wildcard Match – type several characters to select all values that match the given pattern. You can use the asterisk character as a wildcard character. For example, type ca* to select all values that start with the letters “ca.”
o Type In – type explicit dimension member names into a text box to define a filter without querying the database. Use this option when you are using a large data source and queries are slow. If you know the dimension members you are interested in, you can type them into the text box or copy and paste them from another application. Make sure that each member is on its own line in the text box.
o Use All – select all of the members in the data source. Sometimes you will want to define a condition or limit filter that is based on all the data, even if that data changes over time. Rather than selecting specific members to include or exclude from the filter, the Use All option always includes every member in the database as the input to the condition or limit.
• Search Box – When you are working with a field that has a lot of members you may want to search the values and quickly select the ones you are looking for. Type into the search box. Matching values show below the search box. Select the values you want. Each search adds to the selection.
• Selection Controls – These selection controls are available for multidimensional data sources and help you quickly select entire levels. Located at the top of the dialog box, the numbers indicate each level. The color shows what values are selected. The default color means no values are selected for that level, blue means all members on that level are selected, and gray means some members are selected.

• Exclude Mode – By default, selected members when defining a filter will be included and deselected members will be excluded. However, sometimes it is easier to define what you don’t want instead of all of the values you do want show. Select the Exclude option in the upper right corner of the dialog box to make your selections excluded from the filter instead of included.
Adding Conditions to Filters
Use the Condition Tab in the Filter dialog box to define rules to filter by. For example, in a view showing the average Time to Ship for a collection of products, you may want to only show the Products that have a Time to Ship that is greater than 10 days. You can use the built-in controls to write a condition or you can write a custom formula.

Each option on the Condition tab is described below:
• None: select this option if you do not want to add a condition to the filter. This is the default setting.
• By Field: select this option to specify a condition based on existing fields in the data source. Use the first two drop-down lists to select the field and aggregation you want to base the condition on. Then select a condition operator such as greater than, equal to, etc. Finally, type a criteria value into the text box. For example, to create the condition described above, select Time to Ship and AVG from the first two drop-down lists. Then select Greater ( > ) from the operator list and type10 into the text box.
Note:
You can use the Range of Values box to load the entire range of values for the selected field in the data source. The example above would not make sense if all the records in the data source for Time to Ship were greater than 10 days to begin with. Using the Range of Values box helps you decide a value that makes sense to the records in your data source. Click Load to view the range of values for the selected field.
• By Formula: select this option for more advanced filter conditions. You can type a custom formula into the text box or open the formula editing dialog box by clicking the
button to the right of the text box.
Adding Limits to Filters
Use the Top tab in the Filter dialog box to define a formula that computes the data that will be included in the view. For example, in the same view discussed above that shows the average Time to Ship for a collection of products, you can decide to only show the Top 15 Products in terms of Sales. Rather then have to define a specific range for Sales (e.g., greater than $100,000) you can define a limit that is relative to the other members in the field. The formula defined on the Top tab is evaluated on the results of the formula on the Condition tab.

Each option on the Top tab is described below:
• None: select this option if you do not want to add a limit to the filter. This is the default setting.
• By Field: select this option to add a simple limit based on an existing field in the data source. First select the limit range using the first two drop-down lists. For example you can select Top 10 or Bottom 20. Finally select the field and aggregation to base the limit on. So if you wanted to filter based on the Top 10 Sales, select Top and 10 from the first two drop-down lists and then select Sales and SUM from remaining lists.
• By Formula: select this option for more advanced filter limits. Select the limit range using the first two drop-down lists (e.g. Top 10 or Bottom 20). Then you can type a custom formula into the text box or open the formula editing dialog box by clicking the

• button to the right of the text box. For more information on writing formulas and functions refer to .
Example – Filtering Dimensions
This example filters headers and color encodings in a bar chart using the Filter dialog box. To filter the data, follow the steps below.
1. Create the initial data view shown below. It was created using the Superstore Sales Excel data source. The view shows the average regional time to ship for each product based on the container and ship mode.

2. Create a basic filter on the Container dimension that excludes the Small Pack and Wrap Bag shipping containers.
Drag the Container dimension to the Filters shelf to open the Filter dialog box. Click the None button at the bottom of the list to deselect all of the shipping containers. Then select the Exclude option in the upper right corner of the dialog box. Finally, select Small Pack and Wrap Bag. When finished click OK.

3. The view updates to only show orders that were not shipped in a Small Pack or Wrap Bag.

4. Now let’s refine the filter on Container by adding a limit. Right-click the Container field on the Filters shelf and select Filter. The Filter dialog box opens. Leave the selections as they are.
5. Switch to the Top tab and select By Field. Select Top 3 from first two drop-down lists. Then select Sales and SUM from the remaining drop-down lists. When finished click OK.

The Top formula is computed after the selections on the General tab. So first Tableau computes all orders that were not shipped in Small Pack or Wrap Bag containers. Then the view shows just the top 3 of those orders in terms of sales.
6. Now let’s add a new filter on Ship Mode to exclude orders that were shipped via Delivery Truck.
Right-click the Delivery Truck row header and select Exclude. The Delivery Truck ship mode is removed from each region in the view.

7. Finally, let’s filter the Product 2 – Sub-Category dimension to minimize the number of colors being used in the view. Drag the Product 2 – Sub-Category dimension to the Filters shelf.
8. In the Filter dialog box, deselect the Computer Peripherals, Office Machines, and Telephones and Communication values.
The final view is shown below. Take a look at the Filters shelf. You can easily see that the view is filtered on three separate fields. To determine which values have been excluded, open the Filter dialog box for each of these fields.

Filtering Measures
Measures contain quantitative data so filtering this type of field generally involves selecting a range of values that you want to include. There are four types of quantitative filters: Range of Values, At Least, At Most, and Special.
Note:
If you have a large data source, filtering measures can lead to a significant degradation in performance. It is sometimes much more efficient to filter by creating a set containing the measure and then applying a filter to the set.
• Basic Quantitative Filters
• Showing and Hiding Values in the Filter Dialog Box
• Example – Filtering Measures
Basic Quantitative Filters
1. Open the Filter dialog box dragging a measure on any shelf.

2. The Filter Field dialog box opens where you need to specify an aggregation. When finished, click Next.

3. The Filter dialog box opens. There are four types of quantitative filters: Range of Values, At Least, At Most, and Special. Each of these types of filters are described below:
o Range of Values – Specify the minimum and maximum values of the range to include in the view. The values you specify are included in the range.

o At Least – Include all values that are greater than or equal to a specified minimum value. This type of filter is useful when the data changes often so specifying an upper limit may not be possible.

o At Most – Include all values that are less than or equal to a specified maximum value. This type of filter is useful when the data changes often so specifying a lower limit may be not be possible.

o Special – This special type of filter helps you filter on Null values. Include only Null values, Non-null values, or All Values.

4. When finished defining the filter click OK.
5. Showing and Hiding Values in the Filter Dialog Box
6. The filter dialog box shows the minimum and maximum values for the field below the range slider. These numbers give you context when you are deciding the range of values to include in the filter.
7.
8.
9.
10. These minimum and maximum values are affected by the other filters set on the view. For example, a database may include records with sales ranging from $0 to $89K. If you created a filter on the Sales field the minimum and maximum values shown in the filter dialog box would indicate this range. However, let’s say you then filter the view to only show Office Supply products, which sell for between $0 and $25K. By default the filter dialog box will consider that filter and only show the office supplies range. You can use the Show menu in the bottom left corner of the dialog box to switch between Only Relevant Values and All Values in the Database. These options only affect the range that is shown in the filter dialog box and doesn’t change how the filter will be applied to the view.
11.
12.
xample – Filtering Measures
This example filters a text table using an aggregated measure, and then filters the table using the same measure in an unaggregated state.
1. Create the initial view using the Sample – Superstore Sales (Excel) data source. The text table is shown below.

2. Filter the data to only show orders with an average quantity of 26 or more. You can create this type of filter by dragging the Order Quantity measure to the Filters shelf and select Average as the aggregation.
The Filter dialog box is shown below. This type of filter is an At Least filter with the minimum value set to 26.

3. When finished, click OK.
The modified view is shown below. Comparing this view with the original, unfiltered view is straightforward because the measure and the filter use the same aggregation. For example, Copiers & Faxes shipped by Express Air and Regular Air are removed from the view because the average order quantity is less than 26, while Copiers & Faxes shipped by Delivery Truck remains in the view because the average order quantity is greater than 26.

4. Now let’s filter the same view using a disaggregated measure. Suppose you want to filter the view using the disaggregated Order Quantity measure. To do this, select Dimension on the context menu of the AVG(Order Quantity) field on the Filters shelf.

The Filter dialog box is shown below. It displays the limits of the individual rows for the Order Quantity measure. Specify a new lower limit of 26.

The filtered data view is shown below. Notice that the numbers are very different from the original, unfiltered view. This is because Tableau excludes each row in the data source that has an order quantity that is less than 26, and then aggregates the remaining rows as an average.

Filtering Dates
Date fields are a special kind of dimension that Tableau often handles differently than standard categorical data. This is especially true when you are creating date filters. Date filters are extremely common and fall into three categories: Relative Date Filters, which show a date range that is relative to a specific day; Range of Date Filters, which show a defined range of discrete dates; and Discrete Date Filters, which show individual dates that you’ve selected from a list.
• Relative Date Filters
• Range of Dates
• Other types of Date Filters
• Discrete Date Filters
• Example – Filtering Dates
Relative Date Filters
A relative date filter lets you define a range of dates that updates based on the date and time you open the view. For example, you may want to see Year to Date sales, all records from the past 30 days, or bugs closed last week. Relative date filters can also be relative to a specific anchor date rather than today. Follow the steps below to create a relative date filter.
1. Drag a date field from the Data window and drop it on the Filters shelf.

2. In the Filter Field dialog box, select Relative to Now and then click Next.

3. The Filter dialog box opens showing the Relative to Now options. Select a unit of time to filter by. For example, to filter to show the last 2 quarters, select Quarter as the time unit.

4. Use the rest of the controls to define the date filter. You can select from a variety of common options including current, previous, and next. By default, the filter is relative to today. To make the filter relative to an alternate date select the Anchor relative to option in the botom left corner the and select the date to anchor to.
The date period includes the current unit of time. For example, selecting Last 2 Quarters will include the current quarter and the previous quarter. Use the preview in the upper right corner to check your filter settings.

5. When finished, click OK.
Range of Dates
Use this type of filter to define a fixed range of dates. For example, you may want to see all orders placed between March 1, 2009 and June 12, 2009. The Range of Dates filter is similar to the Range of Values option when creating. Follow the steps below to create a Range of Dates filter.
1. Drag a date field from the Data window and drop it on the Filters shelf.

2. In the Filter Field dialog box, select Range of Dates and then click Next.

3. The Filter dialog box opens showing the Range of Dates options. Use the slider or the drop-down date controls to select minimum and maximum dates for the range you want to include. The range is inclusive, which means that the minimum and maximum dates are included in the filter.

4. When finished, click OK.
Note:
If the field also includes Time you can select the Show Times option to further refine your filter range.
Other types of Date Filters
You can also filter dates by defining just a Starting Date or and Ending Date. These filters are are useful when you want to define an open ended range.
In addition, you can create Special filters that include only Null dates, Non-null dates, or All dates.
Use the options at the top of the Filter dialog box to define these types of filters.

Discrete Date Filters
Sometimes you may want to filter to include specific individual dates or entire date levels. This type of filter is called a Discrete Date Filter because you are defining discrete values instead of a range. Follow the steps below to create a discrete date filter.
1. Drag a date field from the Data window and drop it on the Filters shelf.

2. In the Filter Field dialog box, select a date level or select Individual dates and then click Next.

3. In the Filter dialog box, select the dates you want to include.

4. When finished, click OK.
Example – Filtering Dates
This example filters a line graph, to show the profit over a specific range of time. The steps are as follows:
1. Create the initial view shown below. It was created using the Superstore Sales Excel data source. Place Order Date on to the Columns shelf and select All Values as the aggregation. Then place Profit onto the Rows shelf.

2. Now let’s filter the view to include only orders that were place between August 2, 2008 and May 1, 2009. To create this filter drag the Order Date field to the Filters shelf and select Range of Dates in the Filter Field dialog box. Then click Next.
The Filter dialog box is shown below. It displays the Order Date limits. Use the drop-down date controls to specify a new lower limit of August 2, 2008 and an upper limit of May 1, 2009.

The filtered view is shown below.

Using Quick Filters
Tableau lets you quickly add and modify filters using Quick Filters. When you turn on a Quick Filter, a smaller representation of the Filter dialog box opens as a new card. From there you can quickly decide what to include in the view.

• Turning on Quick Filters
• Quick Filter Options
• Searching Quick Filters
• Turning on Quick Filters
• A Quick Filter can be turned on for existing filters or for non-filtered fields. To show or hide a quick filter, select Show Quick Filter from the field’s context menu.
•
•
Quick Filter Options
After you’ve turned on a quick filter there are many different options that let control how the filter works and its appearance. You can access these options using the card menu in the upper right corner of the quick filter card. Some options are available for all types of filters and others depend on whether you’re filtering a Categorical field (dimensions) or a Quantitative field (measures). Finally, you can customize how quick filters display on the sheet, in dashboards, or when saved to the web.
• General Quick Filter Options
• Categorical Quick Filter Options
• Quantitative Quick Filter Options
• Customizing Quick Filters
General Quick Filter Options
• Edit – This option opens the main Filter dialog box so you can further refine the filter by adding conditions and limits.
• Clear Filter – Removes the filter from the Filters shelf and removes the quick filter.
• Make Global – Make the filter global, which means it applies to all sheets that use the same data source. Refer to Global Filters to learn more.
• Only Relevant Values – Specifies which values to show in the quick filter. When you select this option other filters are considered and only values that pass these filters are shown. For example, a quick filter on State will only show the Eastern states when a filter on Region is set. You can use the toggle at the top of the quick filter card to switch between this option and the All Values in Database option.
• All Values in Database – Specified which values to show in the quick filter. When you select this option all values in the database are shown regardless of the other filters on the view.
• Edit Title – By default the title of the quick filter is the name of the field being filtered. Use this option to modify the title. Click Reset to return to the default title.

• Hide Card – Hides the quick filter card but does not remove the filter from the Filters shelf.
Categorical Quick Filter Options
• Include Values – The items selected in the quick filter will be included in the view.
• Exclude Values – The items selected in the quick filter will be excluded from the view.
• Multiple Values List – Displays the values in the quick filter as a list of checkboxes where multiple values can be selected.

• Single Value List – Displays the values of the quick filter as a list of radio buttons where only a single value can be selected at a time. An “All” option can be added to the list to let you quickly select all values without switching to a multiple values list.

• Compact List – Displays the values of the quick filter in a drop-down list where only a single value can be selected at a time.

• Slider – Displays the values of the quick filter along the range of a slider. Only a single value can be selected at a time. This option is useful for dimensions that have an implicit order such as dates.

• Wildcard Match – Displays a text box where you can type a few characters. All values that match those characters are automatically selected. You can use the asterisk character as a wildcard character. For example, you can type “tab*” to select all values that begin with the letters “tab”. Pattern Match is not case sensitive.

Quantitative Quick Filter Options
• Range of Values/Dates – shows the filtered values as a pair of sliders that you can adjust to include or exclude more values. Click on the upper and lower limit readouts to enter the values manually.
The darker area inside the slider range is called the data bar. It indicates the range in which data points actually lie in the view. Use this indicator to determine a filter that makes sense for the data in your data source. For example, you may filter the Sales field to only include values between $200,000 and $500,000 but your view only contains values between $250,000 and $320,000. The range of data you can see in the view is indicated by the data bar while the sliders show you the range of the filter.

• At Least/Starting Date- shows a single slider with a fixed minimum value. Use this option to create a filter using an open ended range.

• At Most/Ending Date – shows a slider with a fixed maximum value. Use this option to create a filter using an open ended range.

• Relative to Now – shows a control where you can define a dynamic date range that updates based on when you open the view. The option is only available for filters on continuous date fields.

• Browse Periods – shows common date ranges such as past day, week, month, three months, one year, and five years. This option is only available for filters on continuous date fields.

Customizing Quick Filters
You can control how a quick filter control appears on the sheet, in dashboards, or when . Customize quick filters by selcting Customize on the quick filter card menu.

Then select from the following options:
• Show “All” Value – toggles whether to show the “All” option that displays by default in multiple values and single value lists.
• Show Search Button – toggles whether to show the search button at the top of the quick filter.
• Show Include/Exclude – toggles whether to show the Include Values and Exclude Values commands on the quick filter card menu. When shown, users can switch the quick filter between include and exclude modes.
• Show Filter Types – toggles whether to let users change the type of quick filter is shown. For example, when shown, a user can change a multiple values list to a compact list.
• Show More/Fewer Button – toggles whether to show the More/Fewer button at the top of the quick filter.
• Show Readouts – controls whether the minimum and maximum values are displayed as text above a range of values. The readouts can be used to manually type a new value instead of using the sliders.

• Show Null Controls – shows a drop-down list that lets you control how the filter handles null values. You can select from from the following options:
o Values in Range – the filter only includes values within the specified range.
o Values in Range and Null Values – the filter includes values within the specified range as well as null values.
o Null Values Only – the filter includes only null values.
o Non-Null Values Only – the filter includes only values that are not null.
o All Values – the filter includes all values. Use this option to quickly reset the selected range to include all values.

Searching Quick Filters
Sometimes a categorical quick filter may contain a lot of values. You can use the Search option to quick find and select the values you want. To open the search field, click the Search icon in the upper right corner of the quick filter card. Then start typing you want to select. Matching values that contain the specified characters will show directly below the search field where you can select or deselect them as needed.
By default, search will return all values that contain the search term. You can use the asterisk character as a wildcard to restrict the results to values that begin with or end with the specified characters. For example, searching for “Bl*” will find all values that start with the characters b and l. Search is not case sensitive.

Global Filters
A global filter is a filter that applies to all worksheets in the workbook that are connected to the same data source. For example, you may have a filter that only includes a specific region or product of interest. Rather than adding this filter every time you create a new sheet, you can simply create the filter once and then make it global.
To make a global filter:
• Right-click an existing filter on the filter shelf and select Make Global.

The field is marked with a globe icon and the filter is applied to all worksheets in the workbook. Additionally, the filter is automatically added to any new worksheet you create. Any changes you make to the filter affects all of the worksheets.
At anytime you can make a global filter local again. When you make a filter local, the filter remains on all the worksheets, however, they are no longer tied together and can be deleted or modified on an individual basis.
To make a global filter local:
• Right-click on the global filter on the filter shelf and select Make Local.

The globe icon is removed and the filter can once again be modified individually per worksheet.
Context Filters
By default, all filters that you set in Tableau are computed independently. That is, each filter accesses all rows in your data source without regard to other filters. However, you can set one or more categorical filters as context filters for the view. You can think of a context filter as being an independent filter. Any other filters that you set are defined as dependent filters because they process only the data that passes through the context filter.
You may create a context filter to:
• Improve performance – If you set a lot of filters or have a large data source, the queries can be slow. You can set one or more context filters to improve performance.
• Create a dependent numerical or top N filter – You can set a context filter to include only the data of interest, and then set a numerical or a top N filter.
For example, suppose you’re in charge of breakfast products for a large grocery chain. Your task is to find the top 10 breakfast products by profitability for all stores. If the data source is very large, you can set a context filter to include only breakfast products. Then you can create a top 10 filter by profit as a dependent filter, which would process only the data that passes through the context filter.
Context filters are particularly useful for relational data sources because a temporary table is created. This table is automatically generated by Tableau when you set the context, and acts as a separate (smaller) data source that results in increased performance when you build data views.
Note:
For Excel, Access, and text data sources, the temporary table is created as an Access table.
• Creating a Context Filter
• Example – Context Filters
Creating a Context Filter
To create a context filter, select Add to Context from the context menu of an existing categorical filter. Alternatively, you can select the Analysis > Set Context menu item. The context is computed once to generate the view. All other filters are then computed relative to the context. Context filters:
• Appear at the top of the Filters shelf.
• Are identified by a grey color and the pushpin icon
.
• Cannot be rearranged on the shelf.
As shown below, the Product dimension is set to be the context for a data view. The Customers filter is computed using only the data that passes through Product.

You can modify a context filter by:
• Removing the field from the Filters shelf – If other context filters remain on the shelf, a new context is computed.
• Editing the filter – A new context is computed each time you edit a context filter.
• Selecting Remove from Context – The filter remains on the shelf as a standard categorical filter. If other context filters remain on the shelf, a new context is computed.
Example – Context Filters
This example walks you through how to create a context filter. First you’ll filter a view to show the top 10 products by sales. Then you’ll create a context filter on product category so you can see the top 10 furniture products.
1. Use the Sample – Superstore Sales data source to create the initial view shown below. The view shows the sales for all products sorted with the highest sale at the top.

2. Now create a Top 10 filter to just show the top selling products. You can create this filter by dragging theProduct 3- Name field to the Filters shelf. In the filter dialog box, switch to the Top tab and define a filter that is Top 10 by Sum of Sales. Refer to Adding Limits to Filters to learn more about defining a Top N filter.

3. When you click OK, you’ll see that the view is filtered to show the top 10 products in terms of sales.

4. Now, let’s add another filter to only show only furniture products. Drag the Product 1 – Category field to the Filters shelf and select Furniture. When finished, click OK.

5. The view is filtered but instead of 10 products, it now only shows 3. The reason is because by default all filters are evaluated separately and the view shows the union of the results. So this view shows that three of the top 10 overall products are furniture products.

6. To find out what the top 10 furniture products are we need to make the Product 1 – Category filter a context filter. Right-click the field on the Filters shelf and select Add to Context.

7. The filter is marked as a context filter and the view updates to show the top 10 furniture products. Tableau has first evaluated the data source and identified all of the furniture products. Then the Top 10 filter is evaluated on the results of that context.

Calculation Filters
Filters on dimensions that are not used elsewhere in the view are called calculation filters. For these types of filters, Tableau performs a calculation on the selected dimension members, This occurs when:
• The dimension is only on the Filters shelf (not used on other shelves).
• You define the filter to include multiple values.
The calculation icon displays next to the field’s name to indicate this operation.
The calculation matches the aggregation for each measure used in the view.
Consider the view shown below. It consists of the Profit measure aggregated as a summation and the Order Quantity measure aggregated as an average. These measures are displayed with the Ship Mode and Containerdimensions. An external filter that consists of two members of the Order Priority dimension is applied to the data.

Tableau automatically applies the appropriate calculation to the members of the external filter based on the aggregation of each measure. Therefore, a summation is performed for Profit and an average is performed for Order Quantity.
For example, the tooltip shows the data for Jumbo Drums delivered by truck. The average order quantity is 24.3. This number was calculated by averaging the order quantities for all the rows that have an Urgent or High order priority. Similarly, the sum of profit is $114,363. This number was calculated by summing the profit for all the rows that have an Urgent or High order priority.

Sorting, Grouping, and Sets
After you understand the basics of building data views, use sorting, groups, and sets to further refine your views and extract exactly the information you are looking for. This section discusses how to re-order and sort the data in a view, filter out unnecessary rows and columns, group dimension members into higher level categories, and create a set using multiple dimensions to create richer encodings.
• So rti ng – Display your data in ascending or descending order based on other fields or custom formulas using computed sorts. Or you can manually sort your data to display in whatever order you choose.
• Groups – Combine dimension members into higher level categories.
• Sets – Create a custom field based on existing dimensions that can be used to encode the view with multiple dimension members across varying dimension levels.
• Sorting
• Groups
• Sets
Sorting
In Tableau, sorting a data view means arranging dimension members in a specified order. Tableau supports computed sorting and manual sorting.
• Computed Sorting
• Manual Sorting
Computed Sorting
You might want to sort customers by alphabetical order, or sort a product line from lowest sales to highest sales. Both of these sorts are “computed sorts” because they use programmatic rules that you define to sort the field.
• About Computed Sorting
• How to Sort Data (Computed Sorts)
• Example – Sorting a Text Table
• Example – Sorting a Hierarchy
About Computed Sorting
Sorting dimensions in a computed manner follows these rules:
• You can sort any discrete field after it has been placed on a shelf (except the Filters shelf).
• Each dimension that appears on a worksheet can be sorted independently of any other dimension.
• The shelf location of the dimension determines the component of the data view that’s sorted. For example, if the dimension resides on the Columns shelf, the columns of the data view are sorted for that field. If the dimension resides on the Color shelf, the color encodings are sorted.
• Sorts are computed based on the values of the filters and sets in the view. Refer to Groups for more information.
• Sorted fields are identified with bold names.
Continuous fields are automatically sorted from lowest number to highest number (as indicated by the axes) and you cannot manually change the sort. However, you can reverse the order of an axis using field specific formatting.
How to Sort Data (Computed Sorts)
Use the sort dialog box to apply computed sorts to fields in the view.
To apply computed sorts:
1. Open the Sort dialog box.
Right-click on the field that you want to sort and select Sort from the its context menu.

2. Specify the sorting options.
Complete the Sort dialog box by specifying the following criteria:
o Sort order – Displays the sort results in ascending or descending order.
o Sort by – Sort by one of these three options:
 Data source order – the order that the data source naturally orders the data. Generally for relational data sources, this tends to be in alphabetical order.
 Alphabetic – the order of the letters in the alphabet.
 Field – order the data based on the associated values of another field. For example, you could order several products by their total sales values.
When sorting by another field, you must also specify the aggregation function to use.
A typical scenario is to sort one or more dimensions by a measure. For example, the Sort dialog box shown below is configured to sort the members of the Customer Segment field in descending order and by the sum of the Sales measure. The results will be displayed so that the member with the highest sales is displayed first, the member with the second highest sales is displayed second, and so on.

You should keep the following rules in mind when interpreting the sort results:
• Tableau computes the sort across the entire table using the specified criteria. Refer to Exam ple – So rti ng a T ext Ta ble for an example.
• Sorts do not break the dimension hierarchy. Sorted fields are always displayed within the ordered context already set forth by the fields on the Rows and Columns shelves. This means that Tableau will not rearrange any of the headers of the fields that appear before (to the left of) the sorted field.

Example – Sorting a Text Table
Using the Sample – Superstore Sales (Excel) data source, this example sorts the rows and columns of a text table to determine which products and years have the highest average discounts. To create the view, follow the steps below:
1. Place the Order Date dimension on the Columns shelf and the Product Sub-Category dimension on theRows shelf.
Complete the text table by placing Discount on the Text shelf and aggregating the measure as an average (select Measure > Average from the field’s context menu). By default, the table is sorted in alphabetical order.

2. Sort the fields.
Right-click on Order Date field and select Sort. In the Sort dialog box select Descending as the Sort Order and sort by Discount aggregated as an Average. When finished click OK. Then apply the same sort to Product Sub-Category.

The view is shown below. Rubber Bands is the top row in the table because it has the largest average discount across all years, while Telephones and Communications are at the bottom in the table because that category has the smallest average discount across all years. Similarly, 2008 is the left most column because it has the largest average discount for all products, while 2006 is the right most column because it has the smallest average discount for all products.

At first glance, it’s not clear if the data has been correctly sorted. That’s because Tableau computes the sort across the entire table using the specified criteria. By turning grand totals on for both columns and rows, using the Table menu, you can see that the sort was performed correctly.

Example – Sorting a Hierarchy
This example uses a multidimensional data source to sort the rows of a bar chart in order to determine which beverages have the highest sales. To create the view, follow the steps below.
1. Place the Sales measure on the Columns shelf and the Gen2,Product dimension on the Rows shelf.
Drill down one level in the hierarchy to display Gen3,Product.

2. Sort Gen3,Product in ascending order by the Sales measure.
Right-click on Gen3,Product and select Sort from the field’s context menu. In the Sort dialog box select Ascending as the Sort order and sort by the Sales field.

The view is shown below. Notice that the Gen3,Product members are sorted within each parent member. For example, Cola, Diet Cola, and Caffeine Free Cola are sorted only within the Colas level. does not rearrange headers that appear before the sorted field.

3. If you want to order dimension members without regard to its parent, you should remove Gen2,Product from theRows shelf. The sorted data are shown below.

Manual Sorting
Manual sorting allows you to rearrange the order of dimension members in the table by dragging them in an ad-hoc fashion, giving precise control over how items appear next to one another in tables and in legends. It also gives you control over the order in which data is drawn on the screen. This control is useful when comparing specific pieces of data or interpreting overlapping data. Manual sorts can only be applied to discrete fields including a discrete measure.
There are two ways to manually sort the data in a view. You can either select items in the view and use the Sort toolbar buttons or you can drag and drop headers in the view.
• Sorting using the Toolbar
• Sort by Drag and Drop
• Example- Manually Sorting Drawing Order
• Sorting using the Toolbar
• The two sort buttons on the toolbar
•
• manually sort a selection either in ascending or descending order based on the other fields in the view. For example, the view below shows sales by product and market size. When you select the Major Market column, thus selecting all of the products, the quick sort buttons sorts the product field by SUM(Sales), which is the measure in the view.
•
•
•
• An easy way to anticipate how a selection will be sorted is by using the tool tips. Make a selection in the view and hover over the ascending or descending quick sort toolbar buttons to see a description of how the selection will be sorted.
• Using the quick sort buttons creates a manual sort which you can always modify using the sort dialog box. Right-click a sorted field (indicated with bold text) and select Sort to open the Sort dialog box.

Sort by Drag and Drop
1. Select the dimension member you want to move. This can be any dimension member that appears in a row or column header of a table, or in a legend like the color legend.
2. Drag the member to the desired location within that row, column or legend.

Example- Manually Sorting Drawing Order
Changing the drawing order of a field allows you to see obscured data in your views in cases where data of one color or shape obscure data of another color or shape. For instance, if you can’t see red marks in a scatter plot because they are obscured by green marks, you can change the drawing order so that the red points are drawn on top of the green points (and vice versa).
Change the drawing order of a field by re-arranging the order of dimension members in a legend. For instance, if you want to place red items in front of green items in a view, select the red legend entry and move it higher on the list of items shown in the legend. The marks are drawn in the view according to the order in the legend, from bottom to top. Also you can toggle back and forth between layered field items by dragging any one of the fields from top to bottom or from bottom to top.
Sorting the drawing order is not restricted to color legends. You can reorder shape legends as well. If you have multiple valid legends, the drawing order is defined first by shape, then by color. For example, suppose you have both a shape legend and a color legend. If you have a red circle on top of a green square, moving the green above the red in the color legend will not necessarily move the green square on top of the red circle. It depends on the order in the shape legend first. If circles are above squares in the shape legend, no amount of reordering the color legend will get that square on top of the circle. Instead, move the square shape above the circle shape first and then reorder the color legend.

Groups
A group is a group of dimension members that have been combined into higher level categories. For example, if you are working with a view that shows average test scores by major, you may want to group certain majors together to create major categories. English and History may be combined into a group called Liberal Arts Majors while Biology and Physics may be grouped as Science Majors.
• Creating Groups
• Editing an Existing Group
• Finding Members in the Groups Dialog Box
Creating Groups
The most common way to create a group is through the group button on the toolbar. However, you can also create groups by right-clicking a dimension in the Data window and selecting Create Group.
To create a group using the toolbar:
1. Hold the CTRL or Shift key on the keyboard to multi-select dimension members in the view.
2. Click the Group button
on the toolbar.

The selected members are combined into a single member and a new grouped field is added to the Data window. A default member name is automatically constructed using the combined member names.

You can use the grouped field just like any other field in the view, except the grouped field is cannot be used to create calculated fields.
You can add to or remove members from a group by right-clicking the grouped field in the Data window and selectingEdit. In the Edit Group dialog box you can also change the default name of the group and combine fields into new groups. Refer to Editing an Existing Group to learn more.
Note:
You can quickly un-group the dimension members by selecting the group in the view and clicking the Group button on the toolbar.
To create groups from the Data window:
1. Right-click a dimension in the Data window and select Create Group.

2. In the Create Group dialog box, select several members that you want to group. Hold the CTRL key on your keyboard to select multiple members.

3. Click the Group button at the bottom of the dialog box.

The selected members are combined into a single member. A default title is automatically constructed using the combined member names. Rename the group by selecting it in the list and clicking the Rename button at the bottom of the dialog box.

Editing an Existing Group
After you have created a group either using the toolbar or from the Data window, you can add members to the group, change the default member names, as well as change the name of the grouped field using the Edit Group dialog box.
To add members to an existing group:
1. Right-click the grouped field in the Data window and select Edit.

2. In the Edit group dialog box, do one of the following:
o Select one or more members and drag and drop them into the existing group. This method works best if you are working with a dimension that has few members.

o Select one or more members, right-click and select Add To. In the subsequent dialog box, select the group you want to add the selected members to and click OK.

o Select one ore more members and select the group in the Add to drop down list at the top of the dialog box.

3. When finished, click OK.
To rename a group:
1. Right-click the grouped field in the Data window and select Edit.
2. In the Edit Group dialog box, select the grouped members and click the Rename button at the bottom of the dialog box.

3. Type a new name and press Enter on your keyboard.
4. When finished, click OK.
Finding Members in the Groups Dialog Box
When you create groups from a large dimension with many members, use the Find option to quickly select the members you are looking for and add them to an existing group.
To use the find options:
1. Show the find options by clicking the Find button at the bottom of the dialog box.

2. Type all or part of the member name into the text box and select an appropriate result criteria from the drop down list. You can select whether to find members that start with, contain, or are an exact match to the search term.
3. Select a Range to search in. You can select to search all members, or within specific groups.
4. Click Find All to select all the matching members or select Find Next to manually navigate through each of the search results.
5. When you have found and selected the members of interest, you can quickly add them to an existing group by selecting the group from the Add to drop-down list at the top of the dialog box
Sets
Sets are custom fields you create that are based on existing dimensions, and that filter data using one or more criteria. You can create a set from any existing dimension. When you create a set for continuous dates associated with a relational data source , the set will be based on discrete values rather than a continuous range of values.
• About Sets
• How to Create a Set
• Creating Sets Examples
About Sets
The three main uses of a set are:
• Create a subset of the data – Select one or more dimension members that are of interest to you. For example, sort a field and select only cities on the west coast with populations greater than 500,000, or manually select outliers that appear in a scatter plot. Refer to Exa mple – A Set Conta ining a Subset for more information.
• Create unique encodings – Combine dimension members to create unique encodings. For example, create a set that combines market and product, and then color-encode a data view using the combined members. Refer to Example – A Set Contain ing Uni que Encodings for more information.
• Save filters for later use – once you have created a filter, you can save the filter as a set and use it in all of the worksheets in a workbook. This saves you from having to recreate the filter every time you want to use it.
Tableau displays sets in the Sets area of the Data window and labels them with the icon.

You can work with a set just as you would with any other dimension. For example, after placing a set on a shelf, you can filter the members, sort the members, and so on.
Additionally, sets are always treated as a filter. Therefore, when you place a set on a shelf, it is automatically placed on the Filters shelf as well.
Note that if you use a filter and a set that are based on the same dimension, the result is the intersection of the filter and the set or its descendents. For example, the following view filters the Store hierarchy to include only the states and the cities shown below.

If you create a set that includes only California, and then place the set on the Filters shelf, the resulting view will contain only the cities in California. That is, the view results from the intersection of the set and the Store filter.

How to Create a Set
You can create a set in one of the following ways:
The best method for you depends on your data characteristics, analysis needs, and so on. If you want to save the sets you create, you should save your work as a workbook or a bookmark. If you do not save any of your work and exit Tableau, your sets will be lost.
• Create a Set by Selecting Marks
• Create a Set from a Field
• Create a Nest Set
• Create a Set by Selecting Marks
• Create a set by selecting marks if you want to create a subset of your data, and the data of interest can best be identified via the data view. For example, you might select outliers or the top few values from a field that’s been sorted.
• Create the set by manually selecting the desired marks in a data view, and then selecting Create Set from the view’s right-click context menu.
• For example, consider the scatter plot shown below. The view consists of two measures that are color-encoded by a dimension. A collection of data points deemed to be outliers are manually selected for a new set.
•
•
•
• Selecting Create Set from the right-click context menu opens the Create Set From Selection dialog box. You can specify the set name, select one or more set members and copy them to the Windows Clipboard, click on a column header to sort the members, or right-click on a column header to remove the column or to restore the original sort order. Changing the sort order in the dialog box does not change the set definition. You should remove columns that aren’t important to your analysis. This will make header labels easier to read and will improve performance.
• Optionally select the Exclude checkbox in the upper right corner if you want the set to contain all members except the ones you selected.
• Note:
• You can optionally select to add the set to the filters shelf after you create it using the check box in the lower right corner of the dialog box.
• Tableau displays the new set in the Sets area of the Data window.
•
•
•
• When you use the set in a data view, a header is created for each set member. As shown below, the header labels are given by the member names.
•
•
• Create a Set from a Field
• Create a set from a field if you want to create a subset of a specific field.
• Create the set by selecting right-clicking the field in the Data window and selecting Create Set.
•
•
•
• The Create Set dialog box opens. Complete the dialog box by specifying the set name and selecting one or more dimension members. In addition, you can optionally define conditions and Top limits to further define the set.
• Tableau displays the new set in the Sets area of the Data window.
•
•
• Create a Nest Set
• A nest set is a cross product of members from different dimensions. You would create a nest set if you want to encode a data view using multiple dimensions. Refer to Example – A Set Contain ing Uni que Encodings to learn more about this method.
• Create the nest set by selecting multiple dimensions in the Data window and then selecting Create Set from the right-click context menu of a selected field.
• For example, the selections shown below will produce a new set that consists of the City and Education Leveldimensions.
•
•
•
• The Create Set From Selection dialog box opens. You can specify the set name, select one or more set members and copy them to the Windows Clipboard, click on a column header to sort the members, or right-click on a column header to remove the column or to restore the original sort order.
•
•
•
• Tableau displays the new set in the Sets area of the Data window.
•
•
•
• When you use the set in a data view, a header is created for each member. The header label is given by combining the original dimension names as shown below.
•
•
section contains the following examples to help you understand how to create and use sets:
• Exa mple – A Set Conta ining a Subset
• Example – A Set Contain ing Uni que Encodings
• Example – Hier archical Sets and their Descendents
• Example – A Set Containing a Subset
• Example – A Set Containing Unique Encodings
• Example – Hierarchical Sets and their Descendents
Example – A Set Containing a Subset
One reason to create a set is so you can easily work with just the dimension members that are of interest to you. For example, you might want to work with specific geographic regions, high-value customers, or one product line in your organization. To create such a set, select the relevant dimension members using any of the methods described in Ho w to Create a Set.
In this example, you will create a subset of the Sample Superstore data source using the Create Set dialog box. Follow the steps below:
1. Select the dimension that will form the set.
Right-click Product 2 – Sub-Category in the Data window, and select Create Set.

2. In the Create Set dialog box, specify the name of the set and select the dimension members that you want to include in the set. In this example, you are only interested in Envelops, Labels, Paper, Pens and Art Supplies, and Rubber Bands.

The new set displays in the Sets area of the Data window. You can edit the set, show set members, and so on using the right-click context menu.

You can use the set to create data views just like any other field.

Example – A Set Containing Unique Encodings
Encoding shelves such as Color, Size, and so on accept only one field at a time. Using the original data source fields, you are limited to encoding your data view with the members of only one dimension. By creating a set, you can encode the view with members from different dimensions.
This example uses the Superstore Sales Excel data source to create a set that contains all the members from two different dimensions. The set is used to encode a data view by color, and is then filtered to include only the members of interest. The steps are as follows:
1. Create the set.
Create the set by selecting the Region and Product 1 – Category dimensions in the Data window, and then selecting Create Set from the context menu.

The Create Set From Selection dialog box opens. Call the new set Product by Region.

2. Encode the data view with the new set.
The data view shown below was created by placing the Customer Segment dimension on Columns shelf, placing the Sales measure on the Rows shelf, and color-encoding the data using the new set.
When you place the set on the Color shelf, Tableau separates the marks according to the members in the set, and assigns a unique color to each member. The color legend displays each member name and its color.

3. Filter the set.
Filter the set to include only the dimension members of interest. You can open the Filter dialog box by selectingFilter on the set’s field menu.

For this example, include only the Furniture and Technology products.

The final view is shown below. Note the name of the filtered set is italicized.

Example – Hierarchical Sets and their Descendents
A hierarchical set filters data to the selected members and all of their descendents. For example, a set named Dairy is created from the Product hierarchy. As shown below, it includes only the Dairy product department.

Consider the following view. The Product Category dimension is placed on the Rows shelf and the Store Salesmeasure is placed on the Columns shelf.

If you place the Dairy set on the Filters shelf, you can see that the view is filtered to include only the Dairy product categories.

As shown below, you can drill down into Product Department to reveal the Product Category, Product Subcategory, and Brand Name levels. As these descendents are revealed, row headers are added to the view. This is because a set filter allows you to view the levels of detail contained within the filtered members.

Dates and Times
Dates in Relational Data Sources
For relational data sources, dates and times are automatically placed in the Dimensions area of the Data window and are identified by the icon. For example, the Order Date and Ship Date dimensions from an Excel data source are shown below.

When you place a relational date on a shelf, the field name is automatically modified to reflect the default date level. Tableau defines the default date level to be the level at which there are multiple instances. For example, if the date field includes multiple years, the default level is year. However, if the date field contains data for just one year but includes multiple months, then the default level is month.
If you don’t want Tableau to automatically select a date level and would rather have a date dimension be a continuous field, you can right-click the field in the Data window and select Convert to Continuous. The dimension turns green in the Data window and anytime you use the field it will be continuous. You can easily revert back by selecting Convert to Discrete from the field’s context menu in the Data window. You can also convert a single field to continuous while it is on a shelf by selecting Continuous on its field menu. The field on the shelf turns green but the field in the Data window is still discrete.
• Changing Date Levels
• Fiscal Dates
• Perfect Pivoting with Dates
• Continuous Dates
• Changing Date Levels
• can change the date level using the field’s context menu after dragging it to a shelf.
•
•
•
• When you select a particular level, Tableau asks the data source to perform a computation on the date field. For example, suppose a particular row in your data source has a date entry of 01/23/07. The year is 2007, the quarter is 1 because January falls in the first quarter, and the week number is 4 because January 23rd falls in the fourth week. How the date level is computed depends on your data source because the computation is actually being done by the data source. Therefore, if your data source is configured to use a specific standard to compute week number, Tableau will use the same standard.
• Note that some date levels might not make sense for your relational data source. For example, if the date format does not include time information such as hour, minute, or second, then selecting one of these options will not add any data to your view.
• You can work with dates at varying levels of detail simultaneously. To do so, you can drill into dates by clicking the control. You can also drag date fields to the Rows or Columns shelf multiple times in order to nest them and to drill down into them at varying levels of detail.
• For example, the view shown below drills down into the year level to display the quarter level as well.
•
•
•
• You can display the data by month by selecting Month from the date field’s context menu. This displays the data for each month across all years.
•
•
•
• To display finer granularity, you can select the MMMM YYYY level from the field menu. Tableau displays the dates using the month and the year.
•
•
Fiscal Dates
Occasionally a date field needs to be expressed in terms of its fiscal date equivalent. For instance, calendar years always run from January 1st until December 31st. But an organization’s fiscal year might start on a month other than January. For instance, a company’s fiscal year might run from June 1st in one year through May 31st of the following year. In these cases, it’s helpful to express the Fiscal Year and the Fiscal Quarter and the Fiscal Week Number rather than their calendar equivalents, when using the date field in a view.
To express date fields in fiscal terms, follow these steps:
1. Right-click the date dimension in the Data window and select Fiscal Year Start. This option is only available on fields that are classified as date dimensions.
2. Designate the start of the fiscal year by selecting a month from the subsequent context menu.
Whether a given level of a date dimension is affected by the conversion to a fiscal equivalent depends on the specific case. Consult the following table:
Date Level When Converted to Fiscal
YEAR The YEAR reflects the fiscal year. For instance, the year for the date June 1, 2004 would be shown as FY 2005.
QUARTER The QUARTER reflects the fiscal quarter. For instance, the quarter for the date June 1, 2004 would be shown as Q1.
MONTH No change in behavior. The calendar month is the same as the fiscal month.
DAY No change in behavior. The calendar day is the same as the fiscal day.
HOUR No change in behavior. The calendar hour is the same as the fiscal hour.
MINUTE No change in behavior. The calendar minute is the same as the fiscal minute.
SECOND No change in behavior. The calendar second is the same as the fiscal second.
WEEKNUMBER The WEEKNUMBER reflects the fiscal week number. For instance, the week number for the date June 1, 2004 would be shown as 1.
WEEKDAY No change in behavior. The calendar weekday is the same as the fiscal weekday.
MM/YYYY No change in behavior. This date format always displays calendar dates, even when a fiscal year has been assigned.
M/D/Y This date format always displays Calendar dates, even when a fiscal year has been assigned.
Notice that the only date level that expressly displays the conversion to a fiscal calendar is the YEAR level. Specifically, fiscal years are shown with the FY prefix. This is not true of fiscal quarters or week numbers, however, which are not shown with any special fiscal markings.
Fiscal year designations for any given date dimension are applied to all instances of the field in the Tableau workbook. Fiscal dates can only be applied to dimensions in a relational data source.
Perfect Pivoting with Dates
You can perfect pivot dates by placing different date levels on different worksheet shelves simultaneously. Place the date field on a variety of shelves and then select the desired date level from the fields’ context menus.
For example, the following line chart displays years as column headers and then color-encodes the marks by quarter.

You can separate the marks by month and by quarter as shown below.

Continuous Dates
You can treat a date as a continuous quantity after placing the field on a shelf. You do this by selecting Continuousfrom the field’s context menu. This draws a quantitative axis for the date values. You can then change the displayed date range by double-clicking on the axis and specifying the desired range.
For example, the view below displays the time to ship as a function of a continuous ship date and color-encoded by region. As you can see, the color of the Ship Date field changes from blue to green after it is converted to a continuous quantity.

Treating dates as a continuous quantity is particularly useful when you use Gantt bars or want to see trends using line charts as shown above.
By default, date dimensions are discrete fields for which Tableau automatically selects a date level when it is placed on a shelf. You can make a date dimension continuous by default by right-clicking the field in the Data window and selecting Convert to Continuous. The field turns green and is automatically converted to a continuous field when you drag it to a shelf. To revert to discrete again, right-click the field in the Data window and select Convert to Discrete.
Reference Lines and Bands
A reference line is typically used to mark a specific value or region on an axis. For example, if you are analyzing the monthly sales for several products, you may want to include a reference line at the average sales mark so you can see how each product performed against the average. Alternatively you may want to shade a particular area along the axis. Finally, you may want to use reference lines to specify a distribution. There are three types of reference lines: lines, bands, and distribution.
Tableau lets you add an unlimited number of reference lines. Add reference lines using the Add Reference Line dialog box.

• Types of Reference Lines and Bands
• Adding Reference Lines
• Adding Reference Bands
• Adding Reference Distributions (Bullet Graphs)
• Editing Reference Lines and Bands
• Removing Reference Lines and Bands
Types of Reference Lines and Bands
There are three types of reference lines and bands
• Line – adds a line at a constant or computed value on the axis. Computed values can be based on a specified field.

• Band – shades an area behind the marks in the view between two constant or computed values on the axis.

• Distribution – adds a gradient of shading to indicate the distribution of values along the axis. Distribution can be defined by confidence interval, percentages, percentiles, quantiles, or standard deviation. In addition to the shading, you can add a line to mark a constant or computed value along the axis. This type of reference line is used to create bullet charts.

Note:
Reference lines are not available when the view is a map using online or offline maps.
Adding Reference Lines
You can add a reference line to any continuous axis.
To add a reference line:
1. Right-click on a quantitative axis and select Add Reference Line.

2. In the Add Reference Line dialog box, select Line.

3. In the Add Reference Line dialog box, select one of the following scopes:

4. Select the Value to mark on the axis. You can select from the following options:
o Average – places a line at the average value along the axis.
o Constant- places a line at the specified value on the axis.
o Maximum – places a line at the maximum value.
o Median- places a line at the median value.
o Minimum – places a line at the minimum value.
o Sum – places a line at the SUM of all the values in either the cell, pane, or entire view.
o Total – places a line at the aggregate of all the values in either the cell, pane, or the entire view. This option is particularly useful when computing a weighted average rather than an average of averages. It is also useful when woring with a calculation with a custom aggregation. The total is computed using the underlying data and behaves the same as selecting one of the totals option the Analysis menu.

5. These values can be applied to any of the measures used in the view. For example, in a view showing sales over time, you can add a reference line that marks the average profit. If there are multiple measures in the view, select the measure to use to compute the reference line.
6. Select how you want to label the line. You can select from the following options:
o None –select this option to not include a label for the reference line.
o Value – select this option to include a label that is the corresponding value on the axis.
o Computation – select this option to display an automatic label. The label is based on the computation and the measure that is selected.
o Custom – select this option to type a custom label into the text box. You can use the menu to the right of the text box to insert values such as the computation or the value.

7. Specify Formatting options for the line. You can change the style, thickness, and color.

8. Optionally, add a Fill color Above and Below the line.

Adding Reference Bands
Reference bands are shaded areas behind the marks in the view between two constant or computed values on the axis. You can add reference bands to any continuous axis.
To add a reference band:
1. Right-click on a quantitative axis and select Add Reference Line.

2. In the Add Reference Line dialog box, select Band.

3. Select one of the following scopes:

4. Specify two values to shade between. For each value you can specify the one of the following values and how you want to label it:
o Average – places a line at the average value along the axis.
o Constant- places a line at the specified value on the axis.
o Maximum – places a line at the maximum value.
o Median- places a line at the median value.
o Minimum – places a line at the minimum value.
o Sum – places a line at the SUM of all the values in either the cell, pane, or entier view.
o Total – places a line at the aggregate of all the vlaues in either the cell, pane, or the entire view. This option is particularly useful when computing weighted average rater than an average of averages. It is also useful wehn working with a calculation with a custom aggregation. The total is computed using the underlying data and behaves the same as selecting one of the totals options in the Analysis menu.

5. Each value can be based on any of the measures used in the view. For example, in a view showing sales over time, you can add a reference band that shades between the average profit and maximum profit. If there are multiple measures in the view, select the measure to use to compute the reference line.
6. Format the reference band. You can mark the two values with a line and select the color to shade between them with.

7. When finished, click OK.
Adding Reference Distributions (Bullet Graphs)
Reference distributions are a variation of reference bands. A reference distribution adds a gradient of shading to indicate the distribution of values along the axis. Distributions can be defined by confidence interval, percentages, percentiles, quantiles, or standard deviation. In addition to the shading, you can add a line to mark a constant or computed value along the axis.
• Basic Reference Distributions
• Bullet Graphs
Basic Reference Distributions
To add a reference distribution:
1. Right-click on a quantitative axis and select Add Reference Line.

2. In the Add Reference Line dialog box, select Distribution.

3. Select one of the following scopes:

4. Select the distribution values. You can select from the following options:
o Confidence Interval – shades the interval between which lie the specified percentage of values.
o Percentages – shades the interval between which lie specified percentages of values. Separate multiple percentage values with a comma (e.g., 60%, 80%, 1000%).
o Percentiles – places a line indicating a specified percentile. When you select this option, you must also select the percentage.
o Quantiles – breaks the view into a specified number of tiles using shading and lines. When you select this computation, you must also select the number of tiles.
o Standard Deviation – places lines and shading to indicated the specified number of standard deviations above and below the mean. When you select this option you must specify the factor, which is the number of standard deviations and whether the computation is on a sample or the population.

5. Specify formatting options. You can format the lines (e.g., style, thickness, and color) as well as the fill gradient. Select from a list of predefined gradients. Select Reverse to change the order of shading in the gradient and Symmetric to use a single color instead of a gradient. You can also specify whether to add additional shading above and below the defined distribution.

Bullet Graphs
Reference distributions can also be used to create bullet graphs. A bullet graph is a variation of a bar graph developed to replace dashboard gauges and meters. The bullet graph is generally used to compare a primary measure to one or more other measures in the context of qualitative ranges of performance such as poor, satisfactory, and good. You can create a bullet graph by adding two reference lines: a distribution to indicate the qualitative ranges of performance and a line to indicate the target.
To create a bullet graph:
1. Select two measures in the Data window. These measures will be compared in the bullet graph. For example, budget vs. actual; actual vs. target; etc.

2. Click the Show Me! button in the toolbar.

3. Select Bullet Graph in the Show Me! dialog box.

Two reference lines are added. By default, Tableau adds a reference distribution that is defined as 60% and 80% of the Average of the measure on the Level of Detail shelf. It also adds a reference line that marks the Average of that same measure. The other measure is placed on the Rows shelf.

You can quickly swap the two measures by right-clicking on the continuous axis and selecting Swap Reference Line Fields.

Edit each of the reference lines to change its definition. For example, you may want to add 100%, or draw a line at a constant value.
Editing Reference Lines and Bands
After you’ve added a reference line or band, you can edit the defintion by right-clicking the continuous axis and selecting Edit Reference Line. If there are multiple reference lines or bands in the view, use the additional menu to select the one you want to edit.

When you have multiple reference lines, you may want to change the order they are drawn in the view. You can reorder a reference line by right-clicking the line and selecting Move to Front or Move to Back.

Removing Reference Lines and Bands
You can remove a individual reference line or band or remove them all at once.
To remove an individual reference line:
• Right-click the reference line in the view and select Remove. If you are removing a reference band or distribution that doesn’t include a line, right-click where at the beginng or end of the shaded area. In distributions, you can also right-click where between the different shades in the gradient.

To remove all reference lines:
• Right-click the continuous axis and select Remove All Reference Lines.

Inspecting Data
Once you have created a view, Tableau offers a selection of dynamic data inspection tools that help you isolate the data of interest and then continue to explore and analyze. For example, if you have a dense data view, you can focus on a particular region, select a group of outliers, view the underlying data source rows for each mark, and then view a summary of the selected marks include the average, minimum, and maximum values.
• Select
• Zoom Controls
• Pan
• Undo and Redo
• Drop Lines
• Summary Card
• View Data
• Describing the View
Select
Selecting marks is useful when you want to visually identify a subset of the data view or you want to run an action.
You can select any individual mark by clicking on it. You can select multiple marks by holding down the Ctrl key. You can also drag the cursor to draw a box around the marks you want to select. Finally, you can combine these methods to quickly select all the marks of interest.
Zoom Controls
Tableau has a set of zoom controls that display in the upper left corner of the view. By default, these contols only display when you hover over a map view. You can control when the zoom controls display by selecting View > Zoom Controls and then select one of the following options:
• Automatic – displays when you hover the mouse over map views.
• Show on hover – displays when you hover the moust over all views.
• Hide – never displays.
These settings also apply to the view when it is opened in Tableau Reader or Tableau Server. You must specify a setting for each worksheet.
The zoom controls allow you to zoom in and out, zoom to a specific area, and fix or reset the axes. Each control is described below.

Zoom In and Out
Zooming is useful when you have a lot of data in a view and you want to focus on a specific part of the view without excluding the rest. Click the plus button to zoom in on the view and the minus button to zoom out. If the zoom controls are hidden, double click the view to zoom in and hold down SHIFT and double-click to zoom out.
Area Zoom
Rather then zooming in and out on the entire view, you can select a specific area to zoom to. When you zoom in on an area, the view is enlarged so that the selected area fills the window. Select the Area Zoom button and then click and drag in the view to select the area to zoom. If the zoom controls are hidden, hold down CTRL + SHIFT and then drag the moust to select the area you want to zoom to.
Reset Axes
When you zoom in or out the axes in the view are locked to a specific range. You can quickly reset the view back to the automatic axis range by clicking the Reset Axes button in the zoom controls. This button is also available on the toolbar.
Pan
You can move your view of a table up and down as well as left and right with the pan tool. There are two uses of panning. The first is when you have zoomed in on a view, particularly a map, and want to move the map around to see other marks of interest. The second is when your data view contains many panes, and you want to move quickly from pane to pane.
Use the Pan tool by holding SHIFT and then dragging the cursor across the view.
Undo and Redo
You can perform unlimited undo and redo of your actions. You can undo almost all actions in Tableau by pressing theUndo button on the toolbar. Likewise, you can redo almost all actions by pressing the Redo button on the toolbar.

In this regard, every workbook behaves like a web browser. You can quickly return to a previous view. Or you can browse all the views of a data source that you have created. Tableau saves the undo/redo history across all worksheets until you exit. The history is not saved between sessions.
Drop Lines
Drop lines are most useful for distinguishing marks and calling out their position in the view. For example, in a view that is dense with scatter marks, you can turn on drop lines to show the position of a particular data point. When you add drop lines a line is extended from the marks to one of the axes. You can choose to show drop lines all the time or only when a mark is selected.

To add drop lines to the view:
• Right-click on the pane and select Drop Lines.
By default, drop lines are set to only show when the mark is selected. You can change this setting and specify other options in the Drop Lines dialog box.
To edit drop lines:
1. Right-click on the pane and select Edit Drop Lines to open the Drop Lines dialog box.

2. In the Drop Lines dialog box select an axis to draw the line to, whether to always show the drop lines, and whether to show labels.
3. When finished click OK
4. Summary Card
5. The summary card is a really quick way to view information about a selection or the entire data source. The card shows the SUM, MIN,MAX, and Average for each measure in the view. You can hide or show the Summary Card by selecting it on the View Cards toolbar menu . You can also select View > Cards > Summary.
6. Consider this example, the view below is a scatter plot of profit vs. sales for three different product categories. You can see that the technology category contains high profit and high sales products (the green marks). When you select these marks, the summary card quickly shows you that these products account for $4,334,791 in sales with a minimum sale of $465,729.
7.
8.
View Data
The View Data command lets you display the values for each row in the data source that compose the marks. It also shows you the summary data based on the aggregations in the view. You might want to do view data to verify the aggregated value associated with a mark, or to isolate and export the individual rows associated with data of interest such as outliers.
You can view data for a selection of marks, the fields in the Data window, and when you’re connecting to data.
The view shown below shows the average order quantity for two product dimensions as a bar chart. Suppose you want to view the data for the largest marks in each pane. To do this, select the marks of interest, right-click in the table, and select View Data on the context menu. Alternatively, you can select the Analysis > View Data menu item.

Note:
Viewing data may not return any records if you are using a field that contains floating point values as a dimension. This is due to the precision of the data source.
• Underlying Data
• Summary Data
• Underlying Data
• The underlying data for the selected marks are displayed on the Underlying tab in the View Data dialog box. Notice that the number of rows that compose the underlying data is shown in the upper right of the dialog box.
•
•
•
• You can sort the data by clicking one or more column headers. To restore the original sort order, click the header repeatedly until it is no longer highlighted with a sort arrow.
• By default, the Show all fields check box is cleared. Select this options to show all the columns in the database rather than just the ones placed on shelves (or fields referenced by a calculation placed on a shelf) in the current worksheet.
• If you want to export one or more data source rows, select the data points of interest by selecting the row of interest clicking Copy to copy the selected data to the Windows Clipboard and paste it into a another file.
• Summary Data
• The summarized data is shown on the Summary tab. The summarized data is a text table of the aggregated data for only the fields shown in the view.
•
•
• Describing the View
• Occasionally you may want to succinctly summarize an analysis you have completed on a worksheet. You might then want to remind yourself of what it shows (the filters that are applied, etc.), and finally, you may want to share a summary of the analysis with someone else.
• When you choose View > Describe Sheet, you can view a description of the workbook, data source, fields and layout of the current worksheet. This summary includes the Caption in the first line, but expounds on other important summary information. This information can be copied and exported to other applications using the clipboard.
• Note:
• If you have Trend Lines turned on, the Describe Sheet dialog box includes information about the trend line model, including an anova table. Refer to to learn more about the terms used to describe the model.
•
•