A/B Testing using Python to determine the most effective marketing campaigns.

A fashion retail company is preparing to launch new product to their apparel catalog. Nevertheless, the management hasn't decided which one of three marketing campaigns in order to promote their new product. Therefore, they decide to disperse and introduce its new product in random selected outlets to examine which marketing campaigns has the substantial impact on product sales. So the company uses different approach of marketing campaigns for each outlets and new product is evaluated by assessing the weekly sales on this month (4 effective weeks).

Our data set consists of 548 observations which includes:

• OutletID: Unique identifier for store location. Each location is identified by a number. The total number of stores is 137.
• AgeofOutlets: Age of store in years (1–28). The mean age of a store is 8.5 years.
• Campaigns: One of three promotions that were tested (1, 2, 3).
• Sales in Thousands: Sales amount for a specific LocationID, Promotion and week. The mean sales are 53.5 thousand dollars.
• Market size: there are three types of market size: small, medium and large.
• Week: One of four weeks when the promotions were run (1–4).

1. Pandas
2. Matplotlib
3. Seaborn
4. Scipy