Tableau's Order of Operations or Everyone's Data Can Dance Many Times

Do you have a dashboard that has filters on it, but they don't seem to work? Tableau has an order that it applies certain actions. You may need to make adjustments to your dashboard to get it to filter the way you want it to.


Remember math class and using an arithmetic mnemonic device such as "Please Excuse My Dear Aunt Sally" for the order you should take performing certain actions (Parenthesis, Exponents, Multiplication, Division, Addition, Subtraction)?


Tableau's order for filtering is Extract, Data Source, Context, Dimensions, Measures, Table Calculations. A mnemonic for this is "Everyone's Data Can Dance Many Times." ( A great friend of mine, Richard Koehler, came up with it.)


The below image is from the help.tableau site here. This provides the order of Tableau filters.


One example where you will have an issue with the filter order is with pulling the top 10 data. The link above has a really good example of how you should adjust your dashboard to look at the top 10 and then the rest of your data.


In the next part of this blog I review what each filter is.

Everyone’s


  • What are Extract filters? This is used to filter the extract from the data source. Think Live versus Extract.


Data


  • How do you add a filter to your Data Source? You can add a filter directly to your data source before you add any to your worksheets, dashboards etc. How to link is here.


Can


  • What are Context Filters? These filters are independent of any of your other filters. Will Perkins (another great colleague) mentioned that you can think of these as your where clause in a SQL statement. Tableau has a how to guide here.

<<<<<Here are the Tableau Functions that occur before Dimension Filters >>>>>

  • Sets - You can create a subset of your data by transforming it into a set. (Helpful Link here) If you were pulling a data set with fruit, you could create a set on fruits that were the color red. If you were pulling a data that had region, you could create a set for the West region.

  • Conditional Filters - This is a filter where you can specify a certain condition that occurs in your data. Think Sum of Sales is greater than 5K or between 1K and 5K.


  • Top N - This is a filter that pulls the top or highest value of your data. What if you wanted to know the top 10 products that were sold in 2021? You could use this filter to pull that data. Alternatively you could also find the bottom values of your data.

  • Fixed LOD or Fixed Level of Detail Expressions - These calculations allow you to compute your data at the data source level. An easy example is a fixed expression such as {FIXED : MIN([Order Date])}. This expression will always return the minimum order date in your data set. You are telling Tableau to look at the Order Date column and to pull on the minimum value. Article Link

Dance


  • What are Dimension Filters? Dimensions are your qualitative values, which do not result from measuring or counting. Examples are names, dates, or geographic data.

<<<<<Here are the Tableau Functions that occur before Measure Filters >>>>>

  • Include / Exclude LOD expressions - Include allows you to consider a dimension that isn't in your workbook such as considering the average blood pressure by state. Exclude is the opposite.

  • Data Blending - this is not a method I like to use in Tableau because it can get complicated easily. Tableau queries the data separately and then brings them together for the view. (Resource link) I prefer to do my data blending in an ETL tool such as Alteryx to not tax the visualizations.

Many


  • What are Measure Filters? Measures are the quantitative data that you filter on measuring and counting. This is the opposite of the dimension filters. For more information, refer to this link here.

<<<<<Here are the Tableau Functions that occur before Table Calc Filters >>>>>

  • Forecasts - This feature in Tableau allows you to see how your data can possibly trend in the future. I wrote a previous blog on it here.

  • Clusters - Clusters allow you to gain insight to different groups and how they perform compared to each other. Reference link here.

  • Totals - Including Percent of Total, etc


Times


  • How does a Table Calculation Work? A Table calculation filter is a special filter that you can specify measure to calculation such as Table across or Table down. Examples are here.

<<<<<Here are the Tableau Functions that after before Table Calc Filters >>>>>

  • Trend Lines - You can add several different kinds of trend lines to your visuals, Linear, Logarithmic, Exponential, Polynomial, and Power. An example on how to use this is if you want to see how data is trending in a scatter plot. Reference Link here

  • Reference Lines - I use these when I want to have a constant line for comparison. An example would be last month's sales compared to this month's sales. I use an example in my Tableau public dashboard here. Here is the Tableau how to link here.



149 views0 comments