Tableau · Beginner

Your first dashboard on Tableau Public

3 min read

Why it matters

Tableau appeared in 64% of the Data Analyst job descriptions we read. Many teams at global companies and consultancies use it, and Tableau Public gives you a free, public profile where recruiters can open your work. A published dashboard is stronger portfolio evidence than a screenshot.

The concept

Tableau builds a query from what you drag onto shelves:

  • Dimensions (blue by default) are categories: Country, Description. They split the view.

  • Measures (green) are numbers that get aggregated: SUM(Revenue).

  • Discrete fields make headers; continuous fields make axes.

  • Rows and Columns shelves set the layout; the Marks card sets colour, size, labels and tooltips.

  • Filters limit data; the order matters (extract, data source, context, dimension, measure, table calculation filters).

A worksheet is one view, a dashboard arranges worksheets, and a story steps through dashboards.

Worked example: Online Retail in Tableau Public

  1. Install Tableau Public (link below) and connect to the UCI online_retail_II.xlsx, sheet Year 2010-2011.

  2. On the data source page, add a data source filter so every sheet sees only real sales: exclude invoices starting with C and keep Quantity above 0. Create the filter with this calculated field and keep True:

Is
NOT STARTSWITH([Invoice], "C") AND [Quantity] > 0 AND NOT ISNULL([Customer ID])
  1. Create Revenue:

Revenue
[Quantity] * [Price]
  1. Sheet "Monthly revenue": drag InvoiceDate to Columns, right-click it and choose continuous Month; drag Revenue to Rows. You get a line from December 2010 to December 2011 peaking in November 2011 at about 1.16 million.

  2. Sheet "Countries": Country to Rows, SUM(Revenue) to Columns, sort descending, and filter to the top 10 by revenue.

  3. Sheet "Customers per country": add COUNTD([Customer ID]) as a label. The Netherlands shows 9 customers, EIRE 3.

  4. Dashboard: title with the finding, KPI text at the top, the line chart, the country bars, and use the country sheet as a filter for the line (Dashboard, Actions, Filter).

  5. File, Save to Tableau Public. Add a description with the data source and licence (UCI, CC BY 4.0).

Common mistakes

  • Counting rows with Number of Records when you need distinct customers. Use COUNTD.

  • Dimension filters that remove the context of a top N. A top 10 filter is computed before or after other filters depending on the filter order. Add the other filter to context when needed.

  • Discrete dates for trends. Discrete month headers break the line when a month has no data. Use continuous dates.

  • Default colours for everything. Tableau colours each category by default; use grey plus one highlight.

Interview angle

Tableau interviews ask "discrete versus continuous", "dimension versus measure" and "what is the order of filter operations". Showing your Tableau Public profile and explaining one design choice on it is often more persuasive than any answer.

Practice task

  1. Publish the dashboard above to your Tableau Public profile.

  2. Add a parameter that switches the bar chart between revenue and number of customers.

  3. Share the link with one person and ask what they learned in 10 seconds.

Further reading