Customer Churn Prediction REST API

const API_TOKEN = 'YOUR_API_TOKEN';

async function predictChurn() {
  const response = await fetch(
    'https://api.cernion.de/api/business-intelligence/churn-prediction',
    {
      method: 'POST',
      headers: {
        'Authorization': `Bearer ${API_TOKEN}`,
        'Content-Type': 'application/json'
      },
      body: JSON.stringify({
        customerSegment: 'prosumer',
        region: 'Bayern',
        riskThreshold: 'high',
        limit: 50,
        includeRetentionStrategy: true
      })
    }
  );
  
  const data = await response.json();
  
  console.log(`At-Risk Customers: ${data.data.atRiskCustomers}`);
  console.log(`Total Revenue at Risk: ${data.data.statistics.totalAtRiskRevenue}`);
  
  // High-Priority Customers
  const urgentCases = data.data.customers.filter(
    c => c.retentionStrategy.priority === 'URGENT'
  );
  
  console.log(`\\nUrgent Cases: ${urgentCases.length}`);
  urgentCases.forEach(customer => {
    console.log(`- ${customer.customerId}: Score ${customer.churnScore}, LTV ${customer.ltv}€`);
    console.log(`  Actions: ${customer.retentionStrategy.actions.map(a => a.type).join(', ')}`);
  });
  
  return data;
}

predictChurn();

import requests
import pandas as pd

API_TOKEN = 'YOUR_API_TOKEN'

def predict_churn(segment='prosumer', region='Bayern'):
    """Churn prediction with retention planning"""
    response = requests.post(
        'https://api.cernion.de/api/business-intelligence/churn-prediction',
        headers={
            'Authorization': f'Bearer {API_TOKEN}',
            'Content-Type': 'application/json'
        },
        json={
            'customerSegment': segment,
            'region': region,
            'riskThreshold': 'high',
            'limit': 100,
            'includeRetentionStrategy': True,
            'includeValueSegmentation': True
        }
    )
    
    data = response.json()['data']
    
    # Convert to DataFrame for analysis
    customers_df = pd.DataFrame(data['customers'])
    
    print(f'\\nChurn Prediction Results:')
    print(f'Total Customers: {data["totalCustomers"]}')
    print(f'At Risk: {data["atRiskCustomers"]} ({data["churnRate"]})')
    print(f'Revenue at Risk: {data["statistics"]["totalAtRiskRevenue"]}\\n')
    
    # High-Priority Actions
    urgent = customers_df[
        customers_df['retentionStrategy'].apply(lambda x: x['priority'] == 'URGENT')
    ]
    
    print(f'Urgent Actions Required: {len(urgent)} customers')
    print(f'Average Churn Score: {urgent["churnScore"].mean():.1f}')
    print(f'Total LTV at Risk: {urgent["ltv"].sum():,}€\\n')
    
    # Retention Budget
    total_budget = sum(c['retentionStrategy']['budget'] for c in data['customers'])
    total_ltv = sum(c['ltv'] for c in data['customers'])
    roi = total_ltv / total_budget if total_budget > 0 else 0
    
    print(f'Retention Investment:')
    print(f'Budget Required: {total_budget:,}€')
    print(f'LTV Protected: {total_ltv:,}€')
    print(f'ROI: {roi:.1f}x')
    
    return data

if __name__ == '__main__':
    result = predict_churn(segment='prosumer', region='Bayern')