Community Spring Cleaning week is here! Join your fellow Maveryx in digging through your old posts and marking comments on them as solved. Learn more here!

Engine Works

Under the hood of Alteryx: tips, tricks and how-tos.
RodL
Alteryx Alumni (Retired)

We all know that working with an example is usually much easier than just reading some instructions and trying to build something out “from scratch.” It’s with this thought in mind that our Alteryx Content Engineering team (@NeilR being the primary contributor) has published a Starter Kit that is focused on the process of A/B Testing and makes available an end-to-end example of how you can complete this type of testing using Alteryx. The Kit is designed to provide what you need to “get started” with both understanding what the test process actually is, as well as how to use the specific tools found in Alteryx.

 

For those who don’t know much about A/B Testing, here’s a snippet from the Introduction document that might spark your interest:

 

In business many times, you try to determine if doing something different will make things better or worse. You may ask questions like:

  • Will a new menu help improve sales in my restaurant?
  • Will a different design of a web page make a difference in how many clicks I get?
  • Will lowering (or increasing) the price of a product increase my net profit?
  • Will changing the wording in a marketing email improve the response rate?

You think to yourself…if only I could test this out in an environment where I can compare the results of changing or not changing something, but yet not roll out the changes to my entire company.

A/B testing is a statistical approach to evaluating the answers to these types of questions, but doing so while only affecting a small subset of your market. Instead of changing something across your whole company – which could backfire if it turns out that the change makes business worse, A/B testing involves applying the change to just a portion of your business and using statistical techniques to infer the potential benefits or detriments of applying the change across the rest of the company.

 

Once you download the Kit and the Introduction document from the Alteryx Gallery and import it, you will be able to go through each step of the process needed to design an A/B Test using a use case that is typical to the retail and marketing world:  What will happen if I change the pricing on a product or service?

 

Within the Kit itself a general overview of the process is provided, and then the workflows needed are built out for you by just clicking through. At any point you can dive into any of the individual workflows to investigate what has been created, with each workflow being fully functional.

 

AB Starter Kit Screenshot.png

 

If you have any interest in A/B Analysis specifically, or would just like to see what is possible for Advanced Analytics in Alteryx, you can download the Kit here.

 

And we always appreciate feedback, so after you download and work through the Starter Kit, we’d love to know if it was a helpful way to learn and how we can improve on this. You can post below any feedback you might have (including suggestions for other Starter Kits you’d like to see us do).

Comments
idaho8007
7 - Meteor

The macro in this starter kit does not work.  Could any one help me with this?  Thanks a lot.

macro problem2.JPG

macro wrong for AB starter kit.JPG

StephenW
Alteryx Alumni (Retired)

Hi @idaho8007

 

There has been an update to A/B analysis as part of the Predictive Analytics Starter Kit.

 

Thanks,

Steve