Understanding the Real Problem - Enterprise Complaints Analytics

A couple of weeks ago, my team was asked to come up with a solution for an Enterprise Complaints Platform with Advanced Analytics capability for a Fortune 50 Bank. The initial scope statements were high level requirements like for example, Identification of high risk complaints that were likely to be escalated to regulatory agencies, Complaint Root Cause Analysis, etc.

It quickly became apparent that while the solution did include Advanced Analytics components what was really needed was a repeatable process for Data Discovery and Descriptive Modeling that would provide the Associates with a semi automated way to complete the Root Cause Analysis or provide a way to identify the initial set of categories that could be used as inputs for Advanced Predictive and Prescriptive Analytics.

Furthermore, the data intake channels for the organization had grown from the traditional banking centers, call centers and ATMs to include “big data” channels including social media and mobile channels where a lot of the data was textual and was not necessarily relevant to the transactional data they got from the more traditional channels.

With these challenges in mind, it became apparent that any proposal we made for an Advanced Analytics solution would require considerable pre-processing capability on the input data including Big Data and a workflow capability to provide the “value add” repeatable automated processes that the front line associates were looking for. While this may seem obvious in hindsight, it was a concept that was not immediately apparent when we started the evaluation with the business looking for a One stop solution to meet their Advanced Analytics needs.  

Views: 763


You need to be a member of Data Science Central to add comments!

Join Data Science Central

Comment by Deborah Deng on November 27, 2014 at 10:12am

Agree. Any analytics work requires probably at least 80% of data work. :)

© 2021   TechTarget, Inc.   Powered by

Badges  |  Report an Issue  |  Privacy Policy  |  Terms of Service