The primary objective for any contact center is to enhance the end-user experience while resolving their issues in the least possible time and cost. Voice of the Customer, collected via various sources, plays an important part in understanding and analysing customer needs and key drivers for satisfaction. While surveys and feedback forms help gather voice of customer, nothing reveals more than the actual conversation with the customer service representatives. While recorded customer conversations are the most direct and accurate form of voice of customer, the huge number of recorded calls makes the monitoring, review and analysing process a daunting task for the analysts.
Speech Analytics presents a solution to this problem and it helps obtain critical business intelligence that would otherwise be lost. Speech Analytics, also known as audio mining, is software that converts unstructured conversations into structured output. It can be performed on recorded conversations or in real time, however the solutions for recorded conversations are more accurate and mature compared to real-time applications.
Success Stories: Speech Analytics Implementation
|One of world’s largest airline||Focus on agents with long calls||38% reduction in Average Handle Time (AHT)|
|A major Insurance company||High number of repeat calls, speech analytics used for changes in agent call script||24% reduction in Repeat Calls|
|One of world’s largest web hosting company||Identify best practices for optimizing sales opportunities||15% increase in revenue per agent|
Data Source: NICE Systems
Speech Analytics solutions have three high level uses:
Key word or Key phrase identification: identify and track the usage of specific words or phrases either specified by the client or words and phrases with a high frequency.
Emotion detection: this would indicate if the caller or agent was happy or unhappy during the conversation. This could be related to the satisfaction levels of the caller. In a real time situation, when the system detects a caller who is unhappy or distressed, it may notify the Supervisor to intervene.
Talk Analysis: the software captures the impact of talk and silence patterns during the call such as the agents placing the call on hold.
Table 1: Speech Analytics Market Share based on number of seats as of August 2012
Data Source: DMG Consulting LLC