<< VIEW FULL RESOURCE LIBRARY

Uncomplicating Complex Data Aggregations in Tableau

March 26, 2020

Tableau

Leveraging Table and Level of Detail Calculations

Complex aggregations allow you to ask intricate questions of your data. You can even nest one aggregation into another for richer, more valuable insights – things like trends, ratios and running totals. But folks avoid these types of calculations or don’t know how to use them, instead doing a hack job using simple calcs.

In this webinar recording, we demystify complex aggregations and show you the available options for performing them in Tableau Desktop. Learn how complex data aggregation requirements can be easily addressed in Tableau by leveraging table calculations and level of detail (LOD) calculations.

You’ll also learn

  • How to filter data displayed in a visualization without filtering the underlying data necessary for aggregated result sets
  • Techniques for creating dynamically recursive calculations such as weighted moving averages
  • The purpose of LOD calculations
  • How to leverage LODs to show data at various grains within the same visualization and address other common aggregation challenges

If you’re a Tableau rookie or on your way to becoming a Tableau guru, you’ll want to add aggregate calculations to your toolset. Understanding the options in Tableau Desktop will open unlimited possibilities for comparing metrics across dimensions and highlighting important insights. You’ll improve data clarity, reduce the number of visualizations needed to communicate insights and increase dashboard performance.

▼ TECHNOLOGIES COVERED

Tableau

▼ PRESENTER

September Clementin
Consultant and Trainer
Senturus, Inc.

September is one of our most sought after instructors. Students love her, giving her 4.9 out of 5 stars. Fluent in both Tableau and Cognos, she has over 15 years of hands-on, expert level experience with a variety of BI platforms. September loves mentoring clients and helping them solve their most complicated problems.

▼ PRESENTATION OUTLINE

  • Aggregation—the process of combining data into groups
  • Types of aggregation
    • Count
    • Sum
    • Average
    • Median
    • Minimum
    • Maximum
    • Percentile
    • Standard deviation
  • Default aggregation
    • Aggregation occurs by default at the lowest level of detail present in a table for individual transactions
    • Measures are aggregated based on the dimensions that are present in the table (columns, rows, pages, and Marks card)
  • Level of Detail calculations
    • Level of Detail (LOD) calculations create the ability to override default aggregation
    • LOD calculations use the following three keywords:
    • INCLUDE — AVG ({ INCLUDE [City] : SUM([Sales]) })
    • EXCLUDE — AVG ({ EXCLUDE [City] : SUM([Sales]) })
    • FIXED — AVG ({ FIXED [City] : SUM([Sales]) })
    • INCLUDE expressions are often used to aggregate data at a level less granular than the transaction level
    • EXCLUDE expressions are often used to omit dimensions in the view when aggregating data
    • FIXED expressions explicitly define the dimensions to be used when aggregating data
  • Table calculations
    • Table calculations are computed locally after aggregated results are returned
    • Table calculations allow for second pass aggregation
      • Rank
      • Running Total
      • Percent of Total
      • Percent Difference
      • Moving Average
      • Compound Growth
      • Percentile
    • Table calculations can return values based on row position
      • Index
      • First
      • Last
      • Lookup
      • Previous Value
    • Order of operations