create_sample_data.py

import pandas as pd
import numpy as np
from datetime import datetime, timedelta

np.random.seed(42)

# Generate sample data
n_customers = 1000
regions = ['North America', 'Europe', 'Asia Pacific', 'Latin America', 'Africa', 'Middle East']
channels = ['Organic', 'Paid Search', 'Paid Social', 'Email', 'Referral', 'Direct']

# Region-based pricing tiers
region_pricing = {
    'North America': [29.99, 49.99, 79.99],
    'Europe': [39.99, 59.99, 89.99],
    'Asia Pacific': [19.99, 39.99, 59.99],
    'Latin America': [14.99, 29.99, 49.99],
    'Africa': [9.99, 19.99, 39.99],
    'Middle East': [24.99, 44.99, 74.99]
}

# Channel CAC ranges
channel_cac = {
    'Organic': (30, 80),
    'Paid Search': (100, 250),
    'Paid Social': (80, 200),
    'Email': (50, 120),
    'Referral': (20, 60),
    'Direct': (40, 100)
}

data = []
start_date = datetime(2023, 1, 1)

for i in range(n_customers):
    customer_id = 1001 + i
    signup_date = start_date + timedelta(days=np.random.randint(0, 365))
    region = np.random.choice(regions)
    channel = np.random.choice(channels)

    # Select price tier based on region
    price_tier = np.random.choice(region_pricing[region])

    # Calculate CAC components
    cac_range = channel_cac[channel]
    marketing_spend = np.random.uniform(cac_range[0], cac_range[1])
    sales_spend = marketing_spend * np.random.uniform(0.2, 0.5)

    # Churn probability (higher for lower price tiers)
    if price_tier < 30:
        churn_prob = 0.6
    elif price_tier < 60:
        churn_prob = 0.4
    else:
        churn_prob = 0.2

    # Generate churn month (blank if not churned)
    if np.random.random() < churn_prob:
        churn_month = np.random.randint(1, 13)
    else:
        churn_month = np.nan

    data.append({
        'customer_id': customer_id,
        'signup_date': signup_date.strftime('%Y-%m-%d'),
        'region': region,
        'acquisition_channel': channel,
        'monthly_subscription_fee': price_tier,
        'churn_month': churn_month if not pd.isna(churn_month) else '',
        'marketing_spend': round(marketing_spend, 2),
        'sales_spend': round(sales_spend, 2)
    })

# Create DataFrame and save to CSV
df = pd.DataFrame(data)
df.to_csv('cac_ltv_model.csv', index=False)
print(f"Created sample data with {len(df)} customers")
print(f"Saved to cac_ltv_model.csv")
print("\nSample data:")
print(df.head(10))