1. Zooming out
It's time to get back to it. You'll start by analyzing the demographics of Databel!
2. The 5 analysis steps from the drag-and-drop exercise
If we look back at the 5 data analysis steps from chapter 1, it's easy to see we are only at step 2.
3. The 5 analysis steps from the drag-and-drop exercise
We did optimize some of our visualizations already, so we don't lose too much time on step 3 - finding fitting visualizations to convey a message. Given Tableau is a data viz tool, it actually makes sense to combine step 2 and 3.
4. Insights discovered so far
We already found some key insights. We know the average churn rate is 27%, and that the main reasons why customers churn is related to competitors. This could raise questions such as "Is the Databel offer competitive enough?". We also discovered the churn rate in California is abnormally high with 63.24%, but it does sound a bit too early to make any general conclusions.
We don't have a clear explanation yet for the relatively high churn rate of 27%, and there are still so many columns we need to analyze.
5. There are many things we don't know yet.
We started creating calculated fields from the get-go to measure the churn rate. This was a great start, but now we're a bit deeper in the analysis we should make sure we have an holistic analysis plan. Looking at the metadata sheet it's easy to realize we only used a handful out of the 29 columns in the database so far. It's a good practice to have a structured approach to your analysis.
6. The metadata sheet is your friend
The metadata in the sheet is grouped in different categories, so let's follow that approach when analyzing the data. We can create different sheets in Tableau to analyze these different topics. Not every analysis will reveal insights, but nonetheless is it important to do your exploratory analysis thoroughly.
7. Let's analyze!
It's time to get back to it. You'll start by analyzing the demographics of Databel! Enjoy!