TMCnet News

CallMiner Gives Comcast Call Analyzing Power
[March 02, 2006]

CallMiner Gives Comcast Call Analyzing Power


By DAVID SIMS

TMCnet CRM Alert Columnist

Fort Myers, Florida-based CallMiner, a developer of enterprise-class speech analytics software, has announced that Comcast Corporation, a provider of cable, internet, and IP-enabled phone service, is deploying CallMiner's Virtual Server Room distributed processing platform to analyze all recorded calls in their Midwest Call Center.



CallMiner's VSR makes possible medium and large speech analytics installations, which were once considered too cost and resource-intensive to deploy, company officials say.

Comcast's 200-seat Midwest Call Center generates thousands of hours of recorded calls each day. A typical speech analytics product would take approximately 20 servers to process Comcast's audio in to minable data. CallMiner's VSR, a patent-pending distributed processing software technology, uses the existing unused computer processing power of Comcast's call center PC's and servers to process all recorded calls for use by CallMiner's Analytics Suite.


The result is that Comcast can now analyze all their recorded calls without having to invest in additional purpose deployed servers to process all the call data.

"Previously, we were not able to effectively listen to the majority of our recorded calls, and we knew there was valuable information that we were missing," said Joe Elias, call center technology manager, Midwest division for Comcast.

He explained that the company was looking for a tool to identify early on "how we can improve customer satisfaction. We can then take that information and make changes that will help us to best service our customers throughout the organization."

Early this year CallMiner announced the award of a five-year GSA contract, GS-35F-0131R, filled to let agencies mine intelligence from all recorded interactions.

CallMiner’s inclusion on the GSA schedule was enabled by an agreement with Carahsoft Technology, a government IT firm based in Reston, Virginia.

The CallMiner Suite uses patent-pending speech recognition technology that converts recorded calls to minable data. The CallMiner Analytics Suite uses data mining and trend mining techniques to analyze recorded interactions that uncover both content and intent embedded in the recordings. This lets call center supervisors mine call content in real time, and hopefully identify call trends and opportunities that, until now, were too costly and time consuming to analyze.

Less than .01% of all agent/customer calls are recorded, and less are mined to uncover what customers are saying, usually simply because it's too costly and IT-resource intensive to do so. A high-performance speech analytics application would require separate dedicated servers to process recorded calls into data suitable for mining, and, well, you can imagine the costs and IT staff needed to maintain all that. In contrast, CallMiner's speech analytics application using VSR would typically only require two servers to perform the same work.

VSR is actually pretty cool, it works by harnessing the unused processing power of desktops in its network. In a typical call center, agents' computers are often using far less than two percent of the computers' processing power, the way people usually use, what, five percent of their brainpower?

VSR finds the unused computing cycles (when the agents stops typing, pauses to speak with the customer or is simply sitting idle) and harnesses that processing power to turn the agent's speech into minable data.

David Sims is contributing editor for TMCnet. For more articles please visit David Sims' columnist page.


[ Back To TMCnet.com's Homepage ]