Abstract | ||
---|---|---|
In many service relationships, customer encounters are not systematically exploited in order to gain valuable insights. However, text mining and analytics methods would provide effective means to systematically screen customer responses and automatically extract relevant business information. In this work, we develop a machine learning method as an artifact for screening incident information in IT Services to detect customer needs. We implement and evaluate the method in a real-world context with an IT provider covering several thousands of incident tickets per year. We show that it is feasible to map incoming tickets to a domain-specific selection of needs-and, hence, enable the providers' customer contacts to address unfilled needs with tailored service offerings. Thus, we contribute a methodology to service marketing and innovation managers to automatically and scalably monitor their customer base for additional sales opportunities. |
Year | DOI | Venue |
---|---|---|
2016 | 10.1109/CBI.2016.30 | 2016 IEEE 18th Conference on Business Informatics (CBI) |
Keywords | Field | DocType |
classification,customer needs,incidents,IT services,machine learning | Customer intelligence,Customer retention,Data science,World Wide Web,Voice of the customer,Customer to customer,Computer science,Customer advocacy,Service level requirement,Customer reference program,Customer Service Assurance | Conference |
Volume | ISSN | ISBN |
01 | 2378-1963 | 978-1-5090-3232-7 |
Citations | PageRank | References |
0 | 0.34 | 0 |
Authors | ||
3 |
Name | Order | Citations | PageRank |
---|---|---|---|
Lena Eckstein | 1 | 0 | 0.34 |
Niklas Kuehl | 2 | 2 | 2.41 |
Gerhard Satzger | 3 | 99 | 23.89 |