We developed a Web-based system for analyzing the voices of call center customers of a life insurance company, so that it would help decision makers understand customer needs better and it would help them make consistent decisions regarding customer support. It used conventional statistical and data mining techniques to identify customer voice patterns. To demonstrate results, we gathered actual customer complaints from the service operation of a target company. Using this data, the system pinpointed problematic areas where complaints happened (one-dimensional analysis), the relationship among problems (two-dimensional analysis), and the root cause of problems (Failure Mode and Effects Analysis). (C) 2004 Elsevier Ltd. All rights reserved.