Skip to content
Tutorials Center
My learning

Topics

Lesson 2 of 35 ยท Welcome to Tableau

Tableau Workspace, Navigation, Design Flow and File Types

Tour the Tableau start page, data pane, shelves, Marks card and tabs, learn the design flow from data to story, and understand twb, twbx, tds, hyper and tflx files.

Before you build charts you need a map of the screen. Tableau has a few places that you will use every day, and a handful of file types that decide how you share your work. This lesson tours both, and then shows a repeatable design flow you can follow for any project.

The start page

When you open Tableau Desktop you see the start page. It has three areas:

  • Connect (left): the list of file types and servers you can connect to, plus saved data sources.
  • Open (middle): recent workbooks and pinned workbooks. Pin the ones you reopen often.
  • Discover (right): sample workbooks, learning links and Tableau Public viz of the day.

Choose a connector and Tableau takes you to the Data Source page. Open a saved workbook and you land on the last sheet you were editing.

The four kinds of tabs

At the bottom of the workbook you see tabs. There are four kinds, and the icons tell them apart.

Tab Purpose
Data Source Connect, join, preview rows, change data types, rename fields
Worksheet One view (a chart, a map or a table)
Dashboard A layout that combines several worksheets and objects
Story A sequence of views and dashboards with captions

Right-click a tab to rename, duplicate, color or delete it. Duplicate is the fastest way to make a variation of a chart. Hide a worksheet that only feeds a dashboard, so users do not see it in a published workbook. Use Ctrl+Z and Ctrl+Y freely: every shelf drag can be undone.

Inside a worksheet

Left side

Data pane Dimensions, measures, sets, parameters
Analytics pane Reference lines, trend lines, forecasts

Drag fields out of the Data pane onto the canvas.

1 / 4
  • The Data pane lists your data source's tables, then fields split into Dimensions (top) and Measures (bottom). It also lists Sets, Parameters and any calculated fields you create.
  • The Analytics pane (a tab next to the Data pane) holds items you drag onto a view: Reference Line, Trend Line, Forecast, Cluster and more. Chapter 6 uses it heavily.
  • The Marks card has a mark-type drop-down (Automatic, Bar, Line, Area, Circle, Shape, Text, Map, Pie, Gantt Bar, Polygon, Density) and buttons for Color, Size, Label, Detail, Tooltip, and sometimes Path, Shape or Angle.
  • The Pages shelf animates a view over a field such as Year. The Filters shelf holds filters, shown later in Filters and the order of operations.
  • The status bar at the bottom shows the number of marks and the sum of selected values, a quick check that a total matches your expectation.

Double-click a field in the Data pane and Tableau places it for you. Double-clicking is fast, but check the result: Tableau may put a pill on Rows when you wanted Columns. You can always drag it across.

The design flow

Good Tableau work follows a simple flow, even when the project is small.

flowchart LR
    A["1. Question<br/>who needs what"] --> B["2. Connect<br/>and check data"]
    B --> C["3. Build<br/>worksheets"]
    C --> D["4. Calculate<br/>and filter"]
    D --> E["5. Assemble<br/>dashboard"]
    E --> F["6. Publish<br/>and iterate"]
    F -.->|feedback| A

The design flow: start from a question, finish with feedback.

  1. Question. Write down who will use this and what decision it supports. A dashboard without a question becomes a wall of charts.
  2. Connect and check. Connect, then inspect rows, data types and null values before building anything.
  3. Build worksheets. One chart per worksheet. Name them well (Sales by Region, not Sheet 7).
  4. Calculate and filter. Add calculated fields, parameters and filters.
  5. Assemble. Arrange worksheets in a dashboard, add interactivity and format for the target device.
  6. Publish and iterate. Share, collect feedback, adjust.

File types

Tableau saves several file types, and the differences matter when you send a file to a colleague.

Extension Name What it contains
.twb Workbook The XML description of sheets and dashboards. No data, only the connection details.
.twbx Packaged workbook A zip file with the .twb plus data files, extracts and images. Opens on any machine.
.tds Data source A shared definition of a connection, joins and calculated fields. No data.
.tdsx Packaged data source A .tds bundled with local data or an extract.
.hyper Extract Tableau's columnar data engine file. (The older .tde format was replaced by .hyper.)
.tfl / .tflx Prep flow A Tableau Prep flow, and its packaged version.
.tbm Bookmark A single worksheet saved for reuse in other workbooks.

Saving rules of thumb:

  • Use .twbx when you share a file that reads a local Excel or CSV file, otherwise the recipient sees "file not found".
  • Use .twb when the data lives on a server that everyone can reach.
  • Publish workbooks to Tableau Cloud or Server for real sharing, so people do not pass files around.

Tableau also keeps a My Tableau Repository folder in your Documents folder. It stores custom shapes, color palettes (Preferences.tps), saved data sources and map sources. Copy it to a new machine to keep your custom assets.

Try it: a packaged workbook is just a zip (stand-in)

A .twbx really is a zip archive that contains a .twb XML file and a Data folder. This Python stand-in builds a tiny fake package in memory and lists what is inside. It is a model of the format, not a real workbook.

Python Stand-in: what a .twbx looks like inside

Press Run (or Ctrl + Enter) to see the result here.

Program input ยท what you'd type while the program runs

Type the answers before you press Run, one per line. Each time the program reads a line (with input()), it gets the next line from this box.

The .twb itself is readable XML. You will rarely edit it by hand, but knowing that a workbook is a text description explains why .twb files are small and why version control tools can diff them.

Python Stand-in: read the XML of a workbook

Press Run (or Ctrl + Enter) to see the result here.

Program input ยท what you'd type while the program runs

Type the answers before you press Run, one per line. Each time the program reads a line (with input()), it gets the next line from this box.

Saving, sharing and versions

A workbook made in a newer version of Tableau cannot be opened in an older one. Tableau offers File > Export As Version to save a copy for an older release. Features that did not exist in that version are dropped, so check the result.

Packaged workbooks grow with the data inside them. If your .twbx is huge, look at whether an extract is bundled (Lesson Connecting to data) and whether you really need all the rows.

Common mistakes

  • Emailing a .twb that reads a local file. The recipient cannot find the data. Package it as .twbx.
  • Leaving default sheet names. In a dashboard with ten worksheets, Sheet 4 helps nobody.
  • Building the dashboard first. Build and test worksheets individually; the dashboard is the last step.

Check your understanding

Quick check ยท 4 questions

of 4 answered You scored / 4 ยท perfect!

  1. 1. Which file type contains the workbook XML AND bundled data files?

    A packaged workbook (.twbx) is a zip with the .twb plus local data, extracts and images. A .twb alone has only the definition.
  2. 2. Where do you drag a Reference Line or Trend Line from?

    The Analytics pane holds items such as Reference Line, Trend Line and Forecast.
  3. 3. What is the first step of the design flow?

    Starting with the question keeps the project focused and tells you which data and charts you need.
  4. 4. Which extension is Tableau's extract format today?

    Extracts use the .hyper format. The older .tde format was replaced by it.

Summary

  • The start page has Connect, Open and Discover; workbooks have Data Source, Worksheet, Dashboard and Story tabs.
  • A worksheet has the Data pane, Analytics pane, shelves, Marks card and Show Me.
  • Follow the design flow: question, connect, build, calculate, assemble, publish.
  • .twb has no data, .twbx packages it, .hyper is the extract, .tds shares a data source definition.

Next: Data types, dimensions and measures