Vizuators' guide

4 steps to Improve Development in Tableau

By Olga Britko, BI developer
April, 2022

You need beauty in visualizations on a dashboard and all that is behind it. In your Tableau documentation, for example. Why and what do we mean by beauty in documentation?


We mean convenience — no need to do extra steps to find what you need.


We mean clearness — the names speak about the content, and the comments suggest the logic of calculations.


We mean speed — it's easy to start with other developer dashboards or were made a long time ago, and now they need improvement.

Here are our 4 cards to examine your dashboard work

Fields from data source
  • Use Comments to fields. Help to understand the specifics of the data.
  • Try not to use Aliases. Not used in calculations, at least. If necessary, create a copy of the field with aliases for signatures, and do the calculations on the original field.
Further work and documentation
  • Folders. Use them to group similarly used fields: filters, formatting, fields with aliases, meaning groups, and numbering.
  • Caption. Brief description of the visualization. It can be used for reference or to make your comments.
  • Describe sheet. Contains the main visualization parameters: Caption, Marks, Filters, Rows, Columns, etc.

It can be used for documentation.

Call:

Worksheet > Describe Sheet

or Ctrl+E (Cmd+E)

Calculated fields
  • Names.
  • Give names your mom can understand.
  • Formulas. Break the logic into steps. Use hyphens to make the formula easier to read. Be careful with duplication. Replace constants with parameters. Create a validation sheet
  • Comments. Describe the purpose of the calculation and the calculation process. Use Describe to Understand the Value of a Field and display metadata.

//single line comment

/* multi-line comment */

Packing work
  • Remove unused fields
  • Use color coding sheets
  • Add update log

Try to avoid similar names of calculated fields

Check out how many similar names we've found on one of the projects we've worked on. A large number of practically identical names generate an error snowball.

word cloud of practically identical names
watch Our video

Rules of development in Tableau (English subs)

Let's try to find all the incorrect names of calculations and pages on your server

GraphQL and Python will help you

How to improve dashboard performance?

Let's audit your dataviz

Use proved Tableau expertise

By clicking the button, you agree to the processing of personal data. We use it to stay connected. Read more