Workforce Optimization Software Featured Article

Combining Call Recording with Speech Analytics Can Improve Multiple Contact Center Operational Metrics

September 26, 2012
By Tracey E. Schelmetic - TMCnet Contributor

Even though the technology has been in existence for multiple decades, there are still many companies who do not fully understand the entire scope and reach of speech analytics. Once perceived to be simply a way to voice-enable the IVR (“Say ‘one’ to be connected to an agent”), speech is so much more than that, and this technology has literally rewritten the way analytics can help a business, particularly a contact center.

As affordable as speech has become, it won’t be too long until speech analytics is a standard, complimentary feature in most call recording systems. Speech already figures prominently as a vital component in workforce optimization suites such as Virtual Observer.

Companies today should consider speech analytics as an affordable tool which helps to solve a number of contact center challenges, including:

  • Enables supervisors to fine-tune agent scripts. Analytics can identify the problem areas with script adherence.
  • Allows for important calls to be discovered, flagged and evaluated or archived. Analytics can help highlight calls in which customers have mentioned a competitor's name, used threatening language, or even to hear when agents speak inappropriately.
  • Altering training programs. With a voice analytics solution, companies can better identify trends in the types of calls the call center is receiving, allowing them to allocate their training resources better. Segmenting into call types can be achieved, and accelerated, using automated call tagging and speech analytics.
  • Increasing first-call resolution (FCR). Analytics can help ferret out which callers have called in about the same issue more than once.
  • Reducing average handle time (AHT). Speech analytics can offer new insights into how agents respond to common questions, allowing the call center to improve responses and shave time off calls.
  • Identifying trends in customer dissatisfaction. By grouping together keyword phrases which may indicate displeasure ("I'm not happy", "not working", "your service sucks"), contact centers can instantly see how many calls match each phrase group.
  • Ensuring compliance. By creating agent-trained trigger phrases such as "credit card number" or "social security number,” call centers can more easily identify calls in which sensitive data may be spoken and recorded. This could have a direct impact on Payment Card Industry (PCI (News - Alert) (News - Alert) (News - Alert)) compliance in particular.
  • Fine tuning marketing campaigns. By creating search categories using product names, brand names, ad slogans and other keyword phrases, a call center or marketing department can better analyze emerging call trends.

Speech Analytics can automate a good amount of your quality assurance processes, freeing up time to accomplish other tasks, launch new initiatives, or reduce resource requirements. Implementing speech analytics in the enterprise requires a bit of trial and error, playing with confidence levels (a confidence level is the percent probability that a positive match will be found when searching for a particular phrase.

For example, a 20 percent confidence level will yield many more results, potentially consisting of a number of false positives, while an 80 percent confidence level will yield fewer results, with the majority being extremely accurate). Most vendors will provide a good amount of experience and insight when implementing a new speech analytics program. The payoffs are indeed tremendous.

Want to learn more about the latest in communications and technology? Then be sure to attend ITEXPO West 2012, taking place Oct. 2-5, in Austin, TX. Stay in touch with everything happening at ITEXPO (News - Alert) (News - Alert). Follow us on Twitter.

Edited by Amanda Ciccatelli

Article comments powered by Disqus