How ‘bot armies’ will transform telco contact centers

A TM Forum Catalyst proof of concept is looking to revolutionize the contact center experience, cutting costs and manual processes for the operator and improving customer experience for the user.

Contact centers represent one of the largest operational expenses for communications service providers (CSPs). To make matters worse, these centers also often deliver poor return on investment in terms of customer experience and revenue. A continuing TM Forum Catalyst project called Cognitive Contact Center aims to revolutionize the contact center experience, cutting costs and manual processes for the operator and improving customer experience for the user.

In the future, customers could interact with contact centers a lot less and have a much better experience when they do, thanks to an army – or ecosystem – of bots working behind the scenes.

Telefónica reaps early rewards

The first phase of the project, which was demonstrated at Digital Transformation World 2018, used a virtual agent, powered by artificial intelligence (AI) and natural language processing (NLP), to make inbound virtual agent interactions with customers much smoother.

The outcomes of this work are already delivering benefits. Telefónica is implementing the tools and best practices generated via the project in its contact centers in several countries, using a virtual agent at the beginning of the customer’s service journey. The virtual agent uses business support systems (BSS) to manage customer interactions. It deploys cognitive technologies, underpinned by NLP, to understand the customer intent and predict the next-best action or operation. When issues can’t be resolved by the virtual assistant, the customer is transferred to a human agent who has a unified service desk for a 360-degree view of all the customers’ interactions so far.

“This is helping us to reduce the number of incoming calls to the call center, and the number of calls that need to be put through to a human agent,” says Francisco Javier García Jiménez, Telefónica. “The work of the Catalyst has been useful for first line resolution and we have seen a good improvement.”

Watch the team discuss the first phase of the project:

Proactive resolution

The latest iteration of the project, Cognitive Contact Center: Phase II, which will be demonstrated at Digital Transformation World in May, expands the scope of this work using data analytics and robotic process automation (RPA) as well as a microservices, event-driven architecture and a multi-bot ecosystem.

Luis Fernando Rubio Martínez, Telecom Director, everis, explains: “In this phase we want to address the end-to-end digitalization of business processes, combining different technologies to reduce manual tasks and to improve customer experience.”

Another important driver is to enable a more proactive approach from the contact center based on predictive analytics and real-time decisions, he says, adding:

“The idea is to gather information in real time both from the network but also from the BSS to anticipate customer problems before they become complaints. Based on machine-learning business rules, we can trigger a proactive virtual agent voice call in real time.”

The virtual agent will try to solve the technical problem online with the customer. As part of this conversational flow, many tasks will be performed behind the scenes, such as retrieval of the customer context profile and transfer to a technical bot to perform diagnosis and service recovery. The team will show how a commercial bot and a technical bot can transfer the interaction between each other while customers continue to interact via a single, simple interface, with all this behind-the-scenes complexity hidden.

If online resolution isn’t possible, a customer visit will be automatically booked through RPA. RPA will also track and manage the service end to end until it is resolved, ensuring the customer is proactively updated.

Further, while this phase demonstrates the interaction of a technical and commercial bot, in the future the concept could incorporate additional specialist bots to help with specific issues such as individual technical areas, billing and more.

Adding value is another focus of this stage, with data analytics detecting the best moment to offer a cross-sell or upsell opportunity to the customer during these proactive interactions.

Tools for the job

The team will use a microservices architecture with TM Forum Open APIs to provide customer context to the virtual agent and enable commercial transactions. The architecture will map to TM Forum’s Open Digital Architecture (ODA), a collaborative vision of a more agile replacement for traditional OSS/BSS architectures. The customer view repository will be mapped against TM Forum’s Information Framework (SID).

The partners have also used TM Forum’s CurateFx digital transformation project management tool to represent business drivers, scenarios and customer journeys. They will use it to map out the interactions and relationships between the different parties involved in the various use cases.

“In this phase we have a complex ecosystem with different parties (customer, RPA, field service team, commercial bot, technical bot, etc.) so we will leverage CurateFx capabilities to guarantee that all interactions required between them are identified and implemented,” Fernando Rubio Martínez explains.

Team roles

Telefónica is the champion of this Catalyst. Catalyst champions bring a business challenge they need to solve and each offers a unique angle based on their geography, strategy and more. They work alongside the Catalyst participants – in this case, everis, Blue Prism, Nokia, Salesforce and Verbio – to develop rapid-fire solutions to the specific challenge over three to six months. Each provides a different piece of the solution puzzle:

• The everis Virtual Agent (eVA) is the core component of the solution, providing an omnichannel and conversational approach to customer interaction management. It serves as the orchestrator of the end-to-end process.


• This phase of the project integrates technology from Verbio for speech-to-text (SST) and text-to-speech (TTS).


• Blueprism provides the RPA technology


• Nokia provides its AI analytics tool, Real Time Triggering, to initiate real-time actions. Nokia’s Service Management Platform executes technical flows to recover customer service failures and includes a bot to diagnose service problems and proactively establish a conversation with the customer to fix the issue or direct them to the correct team.


• Salesforce has joined the project, providing the Service Care platform and the Mulesoft Integration and API Management platform, which deploys several TM Forum Open APIs.

Catalysts are all about the power of collaboration, and the teams deliver outputs which help the entire communications industry to solve shared challenges. In this phase, the team will demonstrate the ODA in action and contribute to the AI & Analytics and Customer Experience use cases. Further, their work will provide extensions to the Information Framework and the TM Forum Open APIs.

Learn more by watching this video filmed at Digital Transformation World 2019: