Wednesday, May 23, 2012

Finding those disgruntled callers...automatically

Contact center supervisors are now using Speech Analytics to automate the process of identifying the calls they really want to listen to.

With some centers recording 10s of 1000s of calls per day, the process of finding the calls was like trying to find the needle in the proverbial haystack.

 The process was made simpler if the contact center agent tagged the call, or if the supervisor knew details about the call - customer name, caller id or agent name, you could then cross-reference a query against those call details, if your system had that information.

Customers now can have all of their recorded calls processed and "listened to".

By creating a category of specific keyword phrases, such as "Disgruntled Callers", the query can be run against the index of analyzed calls and matches can be displayed on screen in seconds.

 What kind of keyword phrases could you use to identify "Disgruntled Callers"?

 "speak to a supervisor"
"mad as hell"
"cancel my service"
"not very happy"
"have to call again"
"need a refund"
"just doesn't work"
"same old runaround"
"find a better deal"
"getting angry"
"HULK SMASH"...and so on.

 There are many different types of speech analytics systems, but they can usually be broken down into two categories: The “Dictionary” based and the “Phonetic” based.

The dictionary-based systems require users to define words and phrases in advance so the speech engine can look for or “spot” those words as the engine is analyzing the audio.

The “Phonetic Based Indexing Engines” are much faster and don't require you to program words in advance.

Our solution, "Virtual Observer Call Insight" is a phonetic-based, enterprise-class Speech Analytics solution that is simple, effective and affordable enough for contact centers of all sizes.

Call Insight provides the ability to receive important search results in real-time within our web-based interface.
Enhanced by Zemanta

Popular Posts