TelecommunicationChurn-February2024.pptx

churn management

1. Churn Prediction Data DoB Usages Amount Recharge Gender Data Usages Other Operator MO Calls Incoming-Outgoing Ratio Monthly Avg Recharge 4G data Usages Movable Subscriber University Areas User Tenure  Special Usages Popular Package subscription SMS MO VOIP suspected VOIP suspected NID

2. Data collected from Source MSISDN TOTAL_MO_DUR_MIN AVG_MO_MIN TOTAL_CALL_COUNT AVG_CALL_COUNT TOTAL_DATA_USAGES_GB AVG_DATA_USAGES_GB TOTAL_MO_SMS AVG_MO_SMS TOTAL_RECHARGE_AMT AVG_RECHARGE_AMT TOTAL_MTC_CALL_COUNT TOTAL_MTC_DUR_MIN AVG_MTC_DUR_MIN LAST_USAGES_DATE STATUS B2W_STATUS AGE GENDER

3. Normalize Data     MO CALL – AVG_DAILY_CALL_COUNT > 1 then 1 – Monthly 30 MO Calls TOTAL_MTC_CALL_COUNT >= 30   AVG_DATA_USAGES_GB > 0.3 - 1 TOTAL_MO_SMS > 1 - 1 TOTAL_RECHARGE_AMT >= 500 

4. Churn Characteristic Based on Gender and Senior-Citizen Female Male Non Senior Citizen Senior Citizen

5. Churn Characteristic Based on Call (MO & MTC) Monthly < 30 Calls > 30 Calls

6. Churn Characteristic Based on DATA & SMS DATA_USAGES_GB > 50 GB Then 1 MO SMS count > 1 then 1

7. Churn Characteristic Based on Recharge Amount Recharge Amount >= 500 TK