Hands-On Data Science for Marketing

Data analysis

Let's take a deeper look at the data. In this section, we are going to focus on understanding the distributions of sales, market sizes, store locations, and store ages used to test different promotions. The goal of this analysis is to make sure the controls and attributes of each of the promotion groups are symmetrically distributed, so that the promotion performances among different groups are comparable to each other.

The total sales distributions across different promotions can be visualized using the following code:

ax = df.groupby(    'Promotion').sum()[    'SalesInThousands'].plot.pie(    figsize=(7, 7),    autopct='%1.0f%%')ax.set_ylabel('')ax.set_title('sales distribution across different promotions')plt.show()

As you can see from ...