My Favorite Tableau 9.0 Feature

• By Joshua Milligan
• Posted in All, Calculations, Level of Detail Calcs, New Features
• On January 14, 2015
• 39 Comments.

Tableau 9.0 has moved from alpha to beta and I’m loving it!  There are many new features to enjoy:

• Improved calculation editor
• Ad-hoc calculations
• New functions (like regular expressions)
• Data interpreter and cleanup tools like pivot and split
• Drag and drop analytics
• Map Search
• New selection options (radial, lasso)
• Responsive tool-tips
• Formatting for Story Points
• And more!

Many of these are incredibly cool and make Tableau that much more smooth and easy to work with.  I’m sure I’ll blog on many of these in the coming weeks.

But there is one new feature that is a true game changer…

(I don’t use the term lightly)

Level of Detail Calculations

(Click here for more posts about LOD calcs.)

Previously, in Tableau, every aggregation was done at the level of detail based on dimensions in the view.  You could roll things up using totals and subtotals.  You could even get a little fancy using sets or table calcs, but these all had their limits.

No longer!

Now you can use Level of Detail aggregations.  Here are some basic examples:

---

 

Fixed: calculates the aggregation at the level of detail specified by the list of regardless of any dimensions in the view.

{FIXED [Department] : AVG([Sales])}

The above code calculates the average sales per Department, regardless what other dimensions are in the view.

Include: calculates the aggregation at the level of detail specified by the dimensions in the view and the dimensions included in the code.

{INCLUDE [Item] : AVG([Sales])}

The above code calculates the average sales at the level of detail defined in the view but includes the dimension Item, even if Item is not in the view.

Exclude: calculates the aggregation at the level of detail specified by the dimensions in the view, excluding any listed in the code.

{EXCLUDE [Category] : AVG([Sales])}

The above code calculates the average sales at the level of detail defined in the view but does not include Category the dimension as part of the level of detail, even if Category is in the view.

---

 

How about a real-life example?

Binning By Aggregate

You may remember that I had previously shared four possible ways of Slicing by Aggregate.  This is when you ask questions of “How many had how many?”

• How many patients had how many readmissions? (e.g. I want to know that 500 patients were readmitted 1 time, 335 were readmitted 2 times, 23 were readmitted 3 times, etc…)
• How many students had how many classes?
• How many products had how many sales?

These types of questions were very difficult (almost impossible) to answer in Tableau without custom scripting or using some crazy options (non-Cartesian visualizations, table calculations, sets) because you have to aggregate the first answer and then use that to slice (or bin) another aggregate.

Let’s say I have a list of students and classes they are taking:

I want to know how many students took how many classes (e.g. 9 students took 6 classes, 5 students took 5 classes, etc…)

In steps Level of Detail (specifically fixed) and solves my problem.  I simply create a calculation like this:

The field shows up as a Measure by default (but I can actually re-characterize fixed level of detail calcs as dimensions!)  Now I just create a bin (right–click the new field and select Create > Bins)

And then I can build the view that answers my question:

This solution doesn’t have any of the drawbacks of the previous work-arounds.  It has nice UI, plays nice with action filters, and isn’t incredibly complex.

And this just barely scratches the surface of what Level of Detail calculations can do!  I’m sure we’ll be seeing a lot more of them in the future.

Want to learn about more Tableau 9 new features?  Check out my other favorite 9.0 features and Data Prep and the Tableau Data Interpreter.