How Australia's largest Telco are measuring and reporting on voice quality using cloud-native tools.
Driving increased CSAT through better audio quality.
The customer engaged Operata to undertake the following:
“Operata proved to be a highly valuable toolset in measuring the performance of our pilot program with Amazon Connect at both a network level and end-to-end - providing key metrics as to the customer and agent experience.”
— Darren Baker, Network Engineering & Planning - Senior Lead, Telstra.
Operata instantly deployed the Operata stats collector into a hosted Amazon Connect Contact Control Panel (CCP). While running test calls through the softphone, the Operata dashboard displayed real-time results, and it was quickly determined that the network was of high quality with all critical metrics in line with industry-standard performance measures.
In reviewing the customers testing methodology, it was determined that their legacy voice testing platform was introducing false negatives, or defects, into the test cases.
This was due to the use of PESQ (Perceptual Evaluation of Speech Quality), a method which was superseded in 2010 by POLQA (Perceptual Objective Listening Quality Analysis). For modern test requirements, the use of POLQA is strongly recommended. The reasoning behind this is because of POLQA’s capability for wideband and super-wideband measurement accuracy and suitability to advanced WebRTC applications such as Amazon Connect.
With the network quickly ruled out as a suspect and the use of the Operata POLQA scoring for Amazon Connect, the customer moved away from an outdated methodology to the Operata cloud-native platform. This captured key 4 metrics across all the right areas:
Now, with the results from these critical measures, Telstra was able to make an informed decision as to the suitability of Amazon Connect to service both agents and customers. On top of that, they now have a tool to continuously measure performance across a large geographic footprint.
Download this Operata Telstra case study PDF.
‍