Customer Cohort Analysis

Cohort analysis is a useful technique to understand customer behavior over time. It involves grouping customers into cohorts based on their first purchase date and then analyzing their behavior over subsequent periods. Below is a Python script using pandas and matplotlib to perform cohort analysis on the provided dataset.

Step 1: Import Libraries

import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
import seaborn as sns
import calendar
import warnings
import json
import os
from datetime import datetime
from itertools import combinations
from collections import Counter

Step 2: Load the Dataset

import pandas as pd
import warnings

warnings.filterwarnings("ignore")

Customer_data_df = pd.read_csv(r"C:\Users\jki\Desktop\Data Scence Projects\Customer Segmentaion Cohort Analysis\Machine Learnign Project\Source Data\customer orders.csv", encoding="ISO-8859-1")
Customer_data_df.head(5)

	Id	Full Name	Address	age	gender	Latitude	Longitude	Email Adress	Order Datetime	Order Status	Order Total	Items	Total sales	Order Count	Rating	Average Rating
0	10	Adam Martinez	China, Beijing Shi	30	Female	40.0	116.0	adam.martinez@internalmail	9/14/2021	CANCELLED	209	Boy's Coat (Blue)	80	1	8	6
1	20	Adam Miller	193, Bannerghatta Main Rd	68	Female	13.0	78.0	adam.miller@internalmail	9/18/2021	COMPLETE	54	Boy's Coat (Blue)	10,816	90	4	6
2	30	Adam Walker	Behrenstraße 42	70	Female	53.0	13.0	adam.walker@internalmail	9/22/2021	COMPLETE	43	Boy's Coat (Blue)	319	3	8	6
3	40	Adan Lamica	Behrenstraße 42	72	Female	53.0	13.0	adan.lamica@internalmail	9/26/2021	COMPLETE	305	Boy's Coat (Brown)	137	3	5	6
4	50	Adeline Iannotti	Floreasca Park 43 Soseaua	16	Female	44.0	26.0	adeline.iannotti@internalmail	9/30/2021	COMPLETE	153	Boy's Coat (Brown)	3,936	64	6	6

Step 3: Convert Order Datetime to Datetime Format

# Convert 'Order Datetime' to datetime format
Customer_data_df['Order Datetime'] = pd.to_datetime(Customer_data_df['Order Datetime'], format='%m/%d/%Y')

Step 4: Extract the Cohort (Month of First Purchase)

# Extract the cohort (month of first purchase) for each customer
Customer_data_df['Cohort'] = Customer_data_df.groupby('Email Adress')['Order Datetime'].transform('min').dt.to_period('M')

Step 5: Calculate the Cohort Index (Months Since First Purchase)

# Calculate the time offset for each order within the cohort
Customer_data_df['Cohort Index'] = (Customer_data_df['Order Datetime'].dt.to_period('M') - Customer_data_df['Cohort']).apply(lambda x: x.n)

Step 6: Group by Cohort and Cohort Index

# Group by Cohort and Cohort Index, then count the number of unique customers
cohort_data = Customer_data_df.groupby(['Cohort', 'Cohort Index'])['Email Adress'].nunique().reset_index()

Step 7: Pivot the Data to Create a Cohort Matrix

# Pivot the data to create a cohort matrix
cohort_pivot = cohort_data.pivot(index='Cohort', columns='Cohort Index', values='Email Adress')

Step 8: Visualize the Cohort Analysis

# Plot the cohort analysis
plt.figure(figsize=(12, 8))
sns.heatmap(cohort_pivot, annot=True, fmt='.0f', cmap='Blues', linewidths=0.5)
plt.title('Cohort Analysis - Customer Retention')
plt.xlabel('Cohort Index (Months since first purchase)')
plt.ylabel('Cohort (Month of first purchase)')
plt.show()

The content you provided appears to be a list of cohorts based on the month of the first purchase, primarily focusing on the years 2022 and 2023. Cohort analysis is a method used to track and analyze the behavior of groups of users over time, often to understand customer retention and engagement.

Here’s a breakdown of the key points:

Cohort Identification: The cohorts are identified by the month of the first purchase. For example, "2022-09" refers to customers who made their first purchase in September 2022.