Customer Potential Revenue Estimation with Rule-Based Classification

Business Problem

A gaming company wants to create level-based new customer personas and estimate how much potential revenue they can generate based on these new customer definitions by using some characteristics of their customers. For example: The company wants to estimate how much revenue an IOS user who is 25 years old and from Turkey can generate on average.

Dataset Story

The Persona.csv dataset contains the prices of products sold by an international gaming company and some demographic information of users who bought these products. The dataset consists of records generated in each sales transaction. This means that the table is not deduplicated. In other words, a user with certain demographic characteristics may have made multiple purchases.

  • Price: The amount spent by the customer
  • Source: The type of device the customer connected from
  • Sex: Gender of the customer
  • Country: The country of the customer
  • Age: Age of the customer

Before

PRICE SOURCE SEX COUNTRY AGE
39 android male bra 17
39 android male bra 17
49 android male bra 17
29 android male tur 17
49 android male tur 17

After

customers_level_based PRICE SEGMENT
0 BRA_ANDROID_FEMALE_0_18 1139.80 A
1 BRA_ANDROID_FEMALE_19_23 1070.60 A
2 BRA_ANDROID_FEMALE_24_30 508.14 A
3 BRA_ANDROID_FEMALE_31_40 233.17 C
4 BRA_ANDROID_FEMALE_41_66 236.67 C