Why Speech Analytics is so Important in Business? The Telefonica Case

Despite disruptive technologies and the large use of web and mobile tools, customer service over the telephone is still the dominant channel of communication between customers and companies. Call centers aim to facilitate these calls and gain whatever insight is detectable from them into customer trends and behavior. Legacy devices and the evolving technologies help the improvement of customer service. Contact centers use technology based on a larger amount of data as data processing is now becoming key to detecting opportunities


There are different solutions that can help call centers to perform better: one of the most powerful is Speech Analytics. This service – through the analysis of recorded call volumes – gathers useful information which can be key to designing a precise picture of the quality and type of interactions between a brand and its customers

Speech Analytics was originally called ‘audio-mining’, a solution which helped audio files to be converted into text to enable searches of specific words or phrases. Nowadays though, Speech Analytics involves in-depth searches and call centers use it to analyze call records for phonemes and to pinpoint key phrases that contribute to a deeper understanding of customer meanings, intentions, and behaviors. This solution is able to detect certain emotions expressed on a phone call as well as trends within a call, such as hold times, silent patches, or agents talking over a caller

‘Implementing a Speech Analytics application solution is not only a farsighted choice but the return of the investments it’s very quick and comes from both significant costs saving and revenue intensification’, says Jose Octavio Fernandes, Country manager Comdata Brazil.


There are already many big companies that are using Speech Analytics solutions such as Telefonica Brazil, the largest telecommunication player in South America which is now taking advantage of Comdata’s solution. ‘We manage an average of 40 million calls every month with IVR (interactive voice response) and we answer by human 9 million calls every month’, said André Santos, Director at Vivo Telefonica Brazil. ‘Those numbers are expected to grow in the future but at the same time the numbers of the human operators will decrease in the next few years, so the challenge is how to increase the quality of our customer experience while reducing the costs. A key solution for us was to implement a Speech Analytics solution’.

‘Comdata’s commitment is, in fact, to maximize the return on investment not only by providing one of the best Speech Analytics solutions on the market but also by helping our clients in optimizing the data and managing the process transformation’, says Octavio Fernandes. More specific Comdata teams will provide a comprehensive analysis of opportunities for using this customer data more effectively, define a target solution for processing and analyzing this data and propose management indicators and new task groups to implement any refocusing required.


Adoption of new customer service technologies has gained speed, especially with Big Data and Machine Learning with Artificial Intelligence (AI) leading the way in intelligent assistants. ‘We decided to implement this solution because the cost for managing and monitoring call center activities is really relevant, and it is increasing year after year. Therefore, it is very important to reduce costs but this without compromising the quality of our service. Those are the main reasons why we decided to use the Comdata solution: it helps increasing the number of monitoring processes in a manner human resources cannot do. Moreover, Speech Analytics helps us to upsurge the quality level of our customer care keeping a higher standard of service’, said André Santos.


There are many benefits that come from the use of a Speech Analytics solution:

  1.  It quickly identifies the reasons for customer calls;
  2.  It reduces call-back activities;
  3.  It reduces the average call handling times;
  4.  It helps call center agents to act on micro and macro customer insight;
  5.  It helps improve the internal skills and services provided;
  6.  It reduces overhead and standard operational expenditures;
  7.  It increases customer loyalty;
  8.  It reduces friction with customers and identification of customers at risk.

​​All these benefits are related to one big goal: improving the customer experience. However, Speech Analytics can also affect the business itself. Such a solution can help identify which sales pitches are generating the best return, and which words or phrases are having the greatest impact on callers. Moreover, call center Speech Analytics is capable of creating personalized cross-sell or up-sell alternatives that meet or exceed a customer’s unique needs.

The transcriptions generated by the Speech services implemented at Telefonica can measure and parameterize multiple situations that occur in potential cross-selling calls. The aim is to estimate the real effect on effective sales to customers in order to design and extend a dictionary of best practices based on data and not on impressions or assumptions. More generally, such service helps evaluating the presence of dissatisfaction, the correct use of the argument, the handling of the different objections, the quality checklist, the presence of insecurity, the doubts presented to the product, and more of 150 possible situations that may happen during a conversation.


If we talk about business opportunity Speech Analytics is an essential support. In Telefonica, we use this solution not only to learn more about our operation but even to acquire best practices of sales to customers. The information we get from this solution is extremely relevant to improve the conversion rate’, said Santos. Through the analysis process, the actions are designed, and their economic impact is estimated for strategic decision making.

The Speech service indeed transcribes 15% of the traffic sampled in a stratified form by agent to obtain a greater scope of vision of the performance of the complete and general service. The objective of this transcription is to measure and parameterize the multiple situations that occur in potential cross-selling calls.


There is another important advantage as a result of implementing a Speech Analytics solution: the reduction of time and management costs through process improvement. Compared to traditional solutions, Speech Analytics allows for evaluation at much greater precision and speed of an enormous quantity of interactions, minimizing human intervention to focus only on the situations that require it.


Join us and let's shape tomorrow together

we make things happen

Please, don't be shy. We would be pleased to help you!

What do you want to talk about ?

Scroll to Top