SHOULD NOT HAVE BEEN IN PROJECT

The project on customers who should not have been customers was very quickly given a high priority. Based on program lead information, we could see that there were huge problems with the way we credit-checked our customers. For example, we found that one man, named Ibrahim, on day number 149 of the year had already been churned 147 times the same year. After some reflection, we did not name the project after him. But here was a man who went out almost once a day, bought a phone for less than $100, and sold it and the SIM card for a personal profit, without ever paying the company. The telecom company at the same time, once a day, chose not to go after the person since it was a relatively small amount in play, and since this person did not have the money anyway. However, the company did not recognize that the same person continuously came back and, under the radar, pulled the same trick. Every time Ibrahim did his trick it would cost the company between $600 and $800 in subsidies to the outlet that sold the phone, subsidies for the telephone, usage of outside networks accessed via the SIM card, and internal processes. And there were quite a lot of Ibrahim’s out there; the trick had become popular.

Since names can be spelled in many ways, we found that searches on an address level provided a better outcome. Exhibit 9.8 shows how the lead information for this program was set up. In this example, you can see from the list that there have been seven terminations ...

Get Business Analytics for Sales and Marketing Managers: How to Compete in the Information Age now with the O’Reilly learning platform.

O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.