The Impatient Analyst Meets Alteryx

Let’s face it – some of us lack the virtue of patience. Having come from a startup background where I spent too many years without a spare second, learning new software for me means diving right in. Add to this impatience the absolute inability to learn from lectures, and you have someone that believes in the “This is Sparta!” approach to learning. Smarter, more patient people take the time to watch videos. Me? I prefer to dive right in.

So, as a business user familiar with Tableau, this is my journey in Alteryx without the videos.

Here’s what I learned the hard way. Somewhere, there is a 10-minute video on all this stuff. If you’re patient, watch the videos and take the proper training. If you, like me, mentally cannot handle 10 minutes of chatter – read this in 60 seconds and go.

The Interface

The Alteryx interface starts out like a cross between Tableau and Microsoft office. Tableau analysts (like myself) will appreciate the lovely blank canvas that encourages you to “drop tools here” and spans most of the working space. It feels a bit like painting with your data.

At the top, below the menu and command shortcuts, experienced Office users will appreciate what I informally call the banner or toolbar. (I’m sure there’s a proper name for this.) To me, this is that huge red toolbox you keep in the garage. Not only do all the tools live in there, but they’re organized into categories. Don’t worry though, because you can bookmark your favorites into the top drawer. Browse tool, you are No. 1 in my book! They even encode these with color and shape so you can begin pattern matching as you learn. Organization and use of pre-attentive attributes make a happy analyst.

To the left, we get the configuration window. This is where the magic happens with each tool. Much like Tableau, these items can be closed or hidden. If you’re like me, you clicked on the “X” here and closed it, much to your dismay. Clicking “View” on the menu and selecting “Configuration Window” will return it, and other objects hyper-clicking causes you to lose. You also input all relevant settings here when you use a tool. (We’ll discuss in more detail in later posts.)

In the lower left, we have an “Overview” window, which acts a bit like a roadmap or viewing pane. (Missing this?  Just like with the configuration, you can enable this at the menu by selecting “View/Overview.”) As these workflows grow and expand, this helps navigate them. Simply grab the box and move it around as you wish. If you want to make your co-workers dizzy, do a demo during lunch and make sure to highlight this feature by spinning about as quickly as possible.

The last piece, the “Results” pane, offers you a great way to preview your data. This, in combination with the “Browse” tool is the secret of the universe. By default, it sits on “Last Run.”  Toggling “Configuration” gives you a neat shortcut to inputs, joins and other parts. You can also drill down into errors, warnings and messages to get an idea of what has occurred.

How It Works

Alteryx is built around the concept of “workflow.” Combining tools replaces scripts visually or, for Excel analysts, all the cut-paste/paste-special and repeat of data cleaning we all dread. (At least, this is my purpose for now.) Other, more advanced users do everything from advanced analytics to spatial file manipulation. My goals, as I start learning, are far less lofty: I just want to clean up some data (in my mind – this is why it’s “Alter Y X,” or Alteryx (“All-tear-ricks” or “All-tricks,” depending on location). The experts tell me this relates back to geocoding.

As Alteryx is visual, I can see the commands I want and select them. It limits classic typos that send us searching for hours (do not look this direction) or the Excel formulas that require Control-Shift-Enter commands or pivot lookups that render handing off our work impossible. It also means that I can copy certain jobs quite easily. The color-coding also helps me match certain tasks and think about what it is I really plan on doing. Once I figure out a particular workflow, re-purposing it is quite easy. Oh, public data pivots and fill-ins, I’m looking at you.

To work with data, one must have data. You’ll generally start with the In/Out area (as a novice), and then work your way through the tools. My favorites so far seem to be In/Out (green file folder), Preparation (blue circle), Join (purple square) and Transform (orangey-red polygon). All the tools use their assigned shapes and colors, making it easier to find related tools. Like Visio, these things will auto-connect if you put them close enough or you can manually connect them. You can also drop them over tools or beside them to add (it displays a nice plus sign). I’m sure there’s a handy keyboard shortcut you can use while dragging to replace.

The Novice’s Favorite Tools

As I’m diving in, my favorite tool lives in the In/Out section: Browse. This lets me review my work in the results pane in a way that is most familiar to me: the actual data. I can see how it looks, what nulls are creating clutter and how many rows came in. I like to put these everywhere as I learn so I can see what’s happening and troubleshoot. You can also click on each and every tool for a quick preview and visualize what the input and output looks like every step of the way. Browse gives you an idea on file size and allows you to save it as a file.

Other tools ideal for Excel and Tableau analysts looking to reshape data:

  • Select Records (preparation area; blue circle with cross hairs) – Great for trimming out random empty rows that Excel sometimes provided (likely, there was data in these before or row 889 has data).
  • Record ID (preparation area; blue circle 1-2-3) – I use this before-and-after transformation to allow some type of tieback and validation.
  • Multi-Row Formula (preparation area; blue circle with a bucket and some drops) – You know those merged cells that give you gray hair? No more! This lets you reference prior cells. Tableau users, consider this like a window calculation.
  • Transpose (transform area; red-orange that looks just like transpose from Excel) – You can select key fields and data fields to un-pivot. Very handy when you need to do multiple pivots within data.
  • Join (join area; purple square with 2 spokes on one side and 3 on the other) – Pick your left and right visually and see what gets excluded.

And for you fellow Spartans, Control-R will run your workflow. Save your mouse-clicks for dragging.

Tune in next time to see a walk-through of my first workflow.

Explore Alteryx Trainings

Share:

Connect

Find out how our team can help you achieve great outcomes.

Insights delivered to your inbox