RSS
Логотип
Баннер в шапке 1
Баннер в шапке 2
Project

Speech analytics of the CST group increased the quality of service in the MTS contact center

Customers: Mobile TeleSystems (MTS)

Moscow; Telecommunications and Communications

Product: SmartLogger II

Project date: 2022/10  - 2023/03

2023: Implementation of Intelligent Voice Analytics for Quality of Service Control at MTS Contact Center

CST Group announced on June 7, 2023, the introduction of intelligent speech analysts to control the quality of service in the contact center. MTS Speech analytics helps to assess the quality of specialist dialogs contact center in 24x7 mode, to move from selective checks to evaluating all dialogs on the stream.

The command CST of integrated voice analytics based on omnicanal the QoS management system and customer interaction analysis. In less than a year, we managed to improve the performance of MTS client service: the number of operator errors decreased by 10%, the level of customer satisfaction increased by 4 percentage points. In addition, the effectiveness of direct increased sales online store by 8%.

it-world.ru
File:Aquote1.png
High quality, speech recognition the ability to aggregate external, data including from, CRM as well as developed neural network algorithms ones help to enrich speech analytics - to implement the most deep analysis. This allows you to understand the true attitude of customers to products and services, quickly respond to changes in needs and structure. market Management receives a summary information in a simple and understandable report, freeing up the resource for strategic tasks, and customers note a fundamentally different level of service,
commented Dmitry Dyrmovsky, CEO of the CST Group of Companies.
File:Aquote2.png

File:Aquote1.png
Thousands of operators work in the distributed contact center of MTS. For effective analytical work, it is necessary to register all dialogs, while it is important to correctly evaluate each appeal. A special requirement was the ability to automatically load and take into account information about a subscriber from CRM - the corporate data warehouse - who he is, whether he is applying for the first time, what services he uses, what personal offer has been formed for him. This allows you to control the supply of a cross-product by employees in 100% of cases, use data to analyze subscriber needs, form comparative samples for customer experience analytics, and much more. According to the results of the pilot project, the voice analytics system of the CST group was chosen, noting the quality of recognition and expanded functionality,
says Georgy Dumikyan, Director of the Customer Service Department, MTS.
File:Aquote2.png

Speech analytics processes text data arrays. To analyze dialogs, the subscriber-operator uses transcribed text, which is obtained as a result of recognition of human spontaneous speech. The CST group uses its own development for this - a technology for speech recognition based on machine learning, as a result of comparison with the solutions of global IT companies and expert universities in the field of speech technologies from the USA, China, the Czech Republic, etc., has been recognized all over the world. The quality of the technologies ensures the conversion of speech into text with the highest accuracy.

Dialogs are analyzed by hundreds of algorithms, instantly highlighting critical points for response, identifying specific specialists for directed training. Analytics allows you to identify the reasons for long-term service, segment responses about competitors and the reasons for the outflow of subscribers. These parameters can be aggregated, which means that service scripts can be improved based on facts, statistics of real calls. Thus, a layer of reliable, daily updated information about subscriber satisfaction and changes in the demand and consumption structure of communication services appears.

Having such information allows you to quickly identify the causes of service failures, change the tariff policy, retain subscribers, and manage the average revenue per subscriber.