Skip to main content

How to Scale Tableau User Filters in 5 Steps

I’ve been working on a sales dashboards where I am visualizing a set of transactional data. One of the requirements was for Tableau Server to recognize the user and their territory upon login. Easy peasy with Tableau User Filters right?

I want to show you how you can do this easily if you have the following complications:

  • Multiple sales reps to 1 region– joining these two datasets will multiply your records (and over-report sales dramatically!).
  • Have a large number of users (I have over 100 reps) and no time to update it manually and constantly as the team changes.
  • Lots of sheets in your workbook and did not want to spend the time adding the user filter set to every sheet.
  • You also dread Tableau’s inevitable “*” error on a blends for your joining table.

Big thanks to the Great Chris Love for this tip!

Step 1: Find Your List of Users and Pivot your Data

You’ll need 2 pieces of data – assigned Tableau Server Usernames, and their respective regions. I’ve mocked up a dummy data set to help illustrate with our favorite retail store – Superstore. Usually the USERNAME() function will be a reflection of your organization’s system username if your Tableau Server is using Active Directory to authenticate/login your users.

As you can see, we have 4 sales regions at Superstore, and 2 sales reps look after each region.

Quick view of my raw dataset

Tableau Usernames for Tableau User Filters

We need to pivot the data so it ends up as a lookup table, but concatenating usersnames to sit in 1 cell. I did this bit in Alteryx.

Alteryx Workflow for Tableau User Filters -Here’s what my workflow looked like with annotations. Alteryx Workflow for Tableau User Filters -Here’s what my workflow looked like with annotations.

BONUS TIP! Did you know you could annotate your tools instead of using the comments box? Much easier when you’re moving around tools in your workflow.

Step 2: Join your lookup table against your transactional data and output your data into Tableau!

Step 3: Connect your output (.tde/.csv) into Tableau and login to your Tableau Server

Connect your data source and sign into Tableau Server. Connect your data source and sign into Tableau Server.

Step 4: Create the following calculation:

Since our usernames are concatenated to exist in 1 cell for every territory, this creates the association between 1 territory and many usernames upon a user’s sign-on.

Step 5: Create a data source filter using the calculation above and Select “True”.
The calculation will look up the username and will filter to the territory specified in your lookup table (now joined against the transaction data)

That was the last step!

You can use the navigation at the bottom to change users views just as you would with a set recommended by Tableau. Hopefully you’ll use this clever method of data pivoting to bring a better UI experience to your Tableau Server users.

If you’d like to download this workbook to see the other calculations, visualisations I have used, you can access the workbook here. Note that the filter will not work with your Tableau Server so I have removed it to avoid confusion.

XML Parsing for Beginners with Alteryx

Recently I’ve come across using XML parsing with web scraping and combining many XML tables. Because of its syntax, I wanted to outline some basic rules about XML and how to work with XML files within Alteryx.

XML files are different from flat columnar tables (the ones we’re used to!) because instead of headers and rows, the data is nested within tags, where the field headers are identified for every record with <> brackets. Below is the example from  Reading XML by Alteryx where the same data is presented as a columnar table and the other in XML. Because of its nested nature, it needs a bit of digging/examining the original data to figure out which tags to target and parse.

Example 1: Parsing from 1 XML file

Let’s go through one recent example I was working on for a client – parsing clinical trial data from XML files. Let’s say I wanted to 2 pieces of information – the name of the trial and when did it start. I like to take a look at the XML file to see where is the information located so I can figure out the best way to parse it.

I’ve opened the file in notepad++ and highlighted the tags which show which pieces of data I am looking for. Note that the root element in this file is “clinical study”, and it just so happens that <brief_title> and <start _date> are child values.

What happens when I select this file in Alteryx’s input tool?

Alteryx’s default settings is to pick up only 1 set of the file’s child values, so there’s a lot of information not yet coming through.

I can bring the information we are trying to parse into Alteryx by selecting “Return Root Element” and  Return Child Values” as the information we are looking for is only 1 layer beneath the Root Element.

Simply add a select tool for the fields we wanted to parse and voila!

But what if I want to parse data that is nested deeper within the XML file?

What if you wanted to find who sponsored the clinical trial?

The data is nested within the <agency> tags, so its not a child value of the root element. This is where can use the Outer XML configuration to bring in all XML tags within the root element. From there, we can parse out the <agency> tags with the ..

XML PARSE TOOL! XML PARSE TOOL!

Simply select “Return Root Element”  and “Return Outer XML” in the input tool so Alteryx can identify both the root element and its respective nested tags.

Then add the XML Parse tool and configure the tool to look for the <agency> tag within the new field “clinical_study_OuterXML”.

I’ve got the results window above with the <agency> data parsed!

XML parsing requires a bit of evaluating the dataset and identifying where your data is located. Because of its nested nature, you might have quite of a few XML parse tools in your workflow. But the same concepts can be taken into web scraping with Alteryx so learning XML and its tagging system has transferable value!

I hope this has been a clear introduction to working with XML files. Let me know if you have any questions in the comments section below.