Marketing Analytics #5: Analyze Away
Do Your Analysis with Confidence!
Do or do not. There is no try.
—Yoda
Before the Good Stuff:
I’ll be posting this shorter post to fill in while I’m taking a vacation.
Expect to get back to regular posts starting next week.
Links to Other Parts
Overview
This post is a short and less prescriptive one than the rest of the Marketing Analytics ones because the act of actual analysis is the most project-dependent.
In this post, I’ll link some relevant resources for analysis based on your project, and show a quick summary of the calculations needed to verify our hypothesis.
Jumping In
By doing the previous steps before taking off into the wilderness, you can ensure that you’ll have a much easier time getting to the treasure than if you took off in a straight line towards the gold.
Data analysis follows the same rules and will be much easier to do once you’ve plotted out a course in advance of jumping immediately into complex math.
In this step, do the appropriate analysis to explore the relationships between various parts of your data. This can take many forms depending on your dataset and goals, but if the previous steps have been completed, this analysis should be straightforward.
This website by NIST is a comprehensive view of different analysis techniques, when to use them, and example data analysis. Look through here to figure out how to best analyze your data!
For our banking dataset, we are analyzing the relationship between deposit rate and contact method. The two options for contact method are landline and cell phone—unordered categorical variables.
To analyze this relationship, we should use a calculation called “Cramer’s V” to determine if there is a meaningful connection.
In our case, we are looking for a value above the accepted minimum cutoff of 0.10 to show a relationship.
With this relationship in hand, we’ve answered “yes” to our original question, “does the contact method affect deposit rate?” Congrats!
Coming Up
In the next post, we’ll go over how to check your answers for common errors and be sure you’re ready to share your result.
If you know anyone who could benefit from learning how to manage their own analytics, feel free to send them a link to the article.
If you want to stay in the loop for future parts, make sure to subscribe to the newsletter.