MTS automated communication with customers of call centers

2021: Automate Contact Center

On April 22, 2021, MTS PJSC announced the completion of the pilot and the transition to scaling the Neuro.net project from the corporate accelerator of the MTS StartUp Hub innovation and investment center. The solution developed by the project will automate the work of the contact center and free its operators from some routine tasks, such as making cold calls or answering standard customer questions.

The voice robot Neuro.net includes the development of NLU (Natural-language understanding), which allows it to learn from a small amount of data. The system operates on the basis of machine learning and knows more than 17 million statements as of April 2021. In addition, the robot is able to make minimal pauses, as in a conversation between a person and a person, and respond to interruption by the interlocutor, silencing and listening to him. The digital operator Neuro.net can make up to 500 thousand calls in 8 hours, which is several thousand times more than the number of calls of a regular operator. According to the results of the pilot project, the Neuro.net decision showed that it is not inferior to people in the conversion of calls.

{{quote 'During the three years of MTS StartUp Hub, 71 pilot projects were launched. Of these, 26 have already switched to scaling. These are startups from a variety of areas, from financial technologies to cloud services. Artificial intelligence is also an important area for MTS. That is why the company decided to introduce the Neuro.net voice robot in its call centers, which showed good results during the pilot, reaching the target KPIs for conversion, "commented Dmitry Kurin, director of MTS Startup Hub. }}

An important feature of our telemarketing scenarios is that we trained digital Neuro.net operators on the most successful calls that the call center previously made as usual. This is the main reason for the success of the pilot project, "said Alexander Kuznetsov, operating director of Neuro.net.