Pricing Optimization using A/B Testing Part II: Additional Insights

Dataset available in Part I

FYI: In my last post I used pivot tables to validate my test group and control group data into information we can actually use.

GOAL: To finish cleaning data for additional insights and prepare for running an A/B Test. This will ultimately allow me to conclude whether the business will benefit from raising the price of their product to $69 or keeping it at $49.

1. Customer Conversion Revenue Data

The first thing we need to decide is which information will ultimately allow us to run the A/B Test we want. Knowing this, it's safe to conclude that returning customers contribute the most to a business's success, so let's focus on these customers. If you're wondering why we are looking at returning customers, it's because they are constant and longterm. If you take a look at the "AB_Test_Results_Verified" sheet you will see that we have our binary assignments for each each customer ID (1=converted, 0=not converted).

Our next step would be to return the revenue for only those customers that converted (1=converted). An IFS statement would be a great option to accomplish this:

2. Additional Insights

Now that we have created our "Revenue" column, we can include that data in pivot tables for additional insights. For instance, let's see what our conversion rate is for our Test Group and Control Group. From these pivot tables we can see that a higher rate of customers converted when they saw the lower price of $49 (control group), but a customer in the test group ($69) yielded a higher average revenue.

CONCLUSION: Now that we have validated, cleaned, and offered additional insights into the data, we can move on to the final step of this project: Concluding whether a price increase will lead to a larger revenue for the company or not.

Coming soon...Performing the A/B Test