TableauPro’s Day 1 with Alteryx – Part 3

This is Part 3 in the series TableauPro’s Day 1 with Alteryx – Intro. If you want to go to Part 2, please click here. In this post, I am talking my impression of the tutorial on Filtering Data. You can watch the corresponding YouTube video here

  • This time, I had to bring 2 Input files and I have brought both of them by drag and drop from Windows Explorer to Alteryx, rather than taking the long route of selecting the tool, and then using the configuration to browse and select the file.
  • I noticed that the recent files are listed on the configuration which is handy.
  • It is pleasing to know that Alteryx can read XML files, though it has been a long time since I have encountered an XML file with data.
  • I have to bring a Select tool to select the columns – I wonder why the Input Data tool itself can’t give this configuration directly to select the needed columns and/or to rename them. Is there a genuine need to seperate Select into a different tool?
  • Here is my first shock with respect to Alteryx. When I look at the datatype next to columns, I see V_String, V_WString, etc. What is the need for these cryptic naming conventions? Also, I realized that I can’t have a whitespace in a name and I am forced to use Underscore in the name. Alteryx does not detect the datatypes too – For Sales, I have to manually point out that it is double – if I am a typical business user (or even an ETL programmer), how in the world am I going to know the exact datatype and its size? I remember using these declaration in the days I was into programming to C++ and Java. I am quite surprised and even disappointed that a company that is saying that it is leading the consumerization of analytics has this kind of a backward interface to one of the most basic operations in the ETL world. Maybe, there is some wisdom in this design choice that I fail to see 🙂 – Can a long-time Alteryx user clarify this for me? Screen Shot 2016-11-07 at 3.09.17 PM.png
  • Here is the next shock – when I read a CSV, every column is read as a string. Then, we need to use a Select tool and manually assign the correct data types. I brought some random CSV file and I tried to detect the right data types and the sizes for the integers and I failed – Alteryx kept telling me that the size is not enough. I had to go through a painful trial and error to get to the right size. If I compare this experience to Tableau, Tableau detects the datatype automatically when a CSV is added as a connection. I understand that the CSV file driver may only output strings, but Alteryx being an App that helps users to do things faster and more productively, why is it not handling it directly?
  • Then, I used the Join tool – here, I was pleasantly surprised by Alteryx behaviour of displaying the L and R (the records that were dropped because they are only in the left side or the right-side respectively). I already liked the T and F in Filter tool and this is super-cool. Often, in Tableau, after doing an inner-join, I try to find out whether any records got dropped by using a number of steps – here, they are all done automatically for me and handed to me in a platter without even me asking for it – Simply Brilliant!Screen Shot 2016-11-07 at 3.23.43 PM.png
  • Then, I added the Summarize tool. Here I expected to double-click and add them to the bottom panel. If I double-click on a text/date field, it can go as a Group By and if I double-click on a numeric field, it can go as SUM(). Typically, this is the behaviour of Excel Pivot Tables for a long time and even Tableau adopts this paradigm. I am surprised that Alteryx has not learnt from Excel/Tableau this simple nice feature to make life easier. Screen Shot 2016-11-07 at 3.28.15 PM.png
  • I often see that there is a red exclamation mark that says there is a problem even when there is no problem. Here, after I have configured the Summarize tool and taken the screenshot, the red exclamation mark was still visible saying that the output was not configured and when I clicked somewhere in the canvas, it disappeared. I see the error message lingering around for longer than it should be, thus causing unnecessary confusion.
  • I now checked out the metadata tab in the Summarize tool to see what it would display as a source. While I am OK with what it is displaying below, I somehow expected to see the original file name where these fields have come from – at least for the Customer Segment, as Summarize is just listing the unique values rather than creating a new values as in the case of SUM_Sales. Tracing out the data lineage from start-to-end in a complex ETL process is often a frustrating problem and I am happy to see what Alteryx has done to help this task. But it could have gone a bit more! Screen Shot 2016-11-07 at 3.31.29 PM.png

This is Part 3 in the series TableauPro’s Day 1 with Alteryx – Intro. If you want to go to Part 4, please click here.

6 thoughts on “TableauPro’s Day 1 with Alteryx – Part 3

  1. Chris says:

    Hey, I am wondering what your main takeaway was from using Alteryx? Is it a product you see yourself or peers using in the future or did it not meet your expectations?

    Like

    • Hi Chris, I have not yet decided. But so far, my impressions have been quite positive, though I do see a number of usability issues with the product as it claims to be a self-service tool like Tableau. But more on them later

      Like

      • mike says:

        Sounds good. Would love to know what your final thoughts are on the product vs. what you’re currently using!?

        Best,
        Chris

        Like

Please give us your feedback...

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s