Creating a Cognitive Digital Agent
22 Dec 2016
Creating a Cognitive Digital Agent
What if you could transform customer service while saving between 10 and 40 percent? And if that were just the beginning? Monetization, new business models, partners from other industries could all be future missions for the cognitive digital agent.
Where did the idea for a Cognitive Digital Agent come from and how did it come into being? The Philippines’ Globe Telecom looked at customers’ challenges when trying to serve themselves – such as negotiating IVR menus and hard to navigate websites – and matched them with the ‘mirror’ image of those challenges – how to fix them and beyond, to provide a very different kind of experience.
Globe recognized that self-care channels are good for customers and service providers alike, if they’re done right. This is because customers like the speed, convenience and 24-hour availability, while service providers want to improve customer service, but need to lower the associated costs.
Globe proposed this difficult issue as the basis for an accelerated, collaborative R&D project within TM Forum’s Catalyst program. It was accepted and Globe became the champion of a project “to explore the use of cognitive computing or artificial intelligence technologies, such as natural language processing and machine learning, to create the next generation of customer care services.”
Globe was supported by participants Amdocs, Huawei and Infosys. One of the key parts of this initiative was the creation of a cognitive digital agent that could answer questions on various channels, including IVR, and only funnel the more complex queries or those that cannot be resolved automatically to human beings.
Clearly a key requirement was that it understands natural human voices and combines artificial intelligence, machine learning and the latest customer relationship management (CRM) techniques to produce answers. Not only does this streamline how each query from a customer is handled, but the machine learning element means that the more queries it deals with, the greater its knowledge base. Further, the agent was intended to make interchanges, which customers used to find exceedingly irritating, fun as well as efficient – no mean feat.
This is easy to say, but the machine learning element, which was the major focus, was also the most difficult. This is new territory and questions like, “What are the best ways to train a cognitive digital agent so that it can genuinely support and improve the customer experience?” are entirely new territory.
The Catalyst was also intended to understand how artificial intelligence and machine learning can augment – rather than replace – the human workforce.
The team used TM Forum’s Open APIs to integrate all the elements of the project and drew on the Forum’s pioneering omnichannel work, which has been tried, tested and progressed in a whole series of Catalysts. The single cognitive computing platform, linked to the CRM, is a big move towards breaking down silos and dramatically speeding up implementations.
Nik Willetts, Deputy CEO, TM Forum, commented, “This Catalyst has the potential to reduce costs by between 10 and 40 percent, and Globe intends to implement it over the coming months” on announcing The Cognitive Digital Agent has won the best-in-show award* (out of the ten Catalyst demonstrations) at the recent TM Forum Live! Asia event.
He added, “This winner showcased the business outcomes exceptionally well, and the inventive use of machine learning and artificial intelligence”.
What did Globe make of its first Catalyst experience and the outcome? We asked Vincent Seet, Head of Enterprise Architecture, Globe Telecom. He said, "It has been an exhilarating journey over the last four months… [Once the value proposition was understood] we had the sponsorships from the senior leaders in our organizations and the project team went out full force to deliver the vision of the Catalyst.”
He added, “The project team has certainly learned a lot from this Catalyst project. It has become clear to us that the Catalyst program is a very effective platform to encourage innovation and solve problems faced by the community. The collaborative model is something that service providers would not be able to replicate on their own without always having commercial obligations.
“Apart from that, TM Forum has taken care of the contractual aspects in the collaboration, which allowed the champions and participants to truly focus on solving the problem at hand. The TM Forum community should really leverage this platform to its fullest potential."
Looking to the next steps, Richard Im, Director of Carrier Software, Huawei, said, “Going forward it’s more [about how] omnichannel experiences and different kinds of analytics for understanding customers’ behavior could connect with what we’re doing. That will give us a lot of information to see what the business can do and acting on it to monetize, but also to retain customers.
“It could be used to develop different business models with other industries in future, combining those elements together. So, this is a great starting point and from here we could move into all sorts of areas, including collaborating with other Catalyst projects – we have a vision of getting together with other Catalysts and coming out with new, innovative ways to work together.”
Watch the video interview with Vincent and Richard here and stay tuned to Inform to see what happens next.
* The judging panel judging panel comprised Jenny Huang, Lead of OSS/BSS Standards Strategy Group, AT&T; David Pleasance, Chairman, TM Forum; Michael Lawrey, Board Member, TM Forum; and Nik Willetts, Deputy CEO, TM Forum. Check out the other winners here.
Where did the idea for a Cognitive Digital Agent come from and how did it come into being? The Philippines’ Globe Telecom looked at customers’ challenges when trying to serve themselves – such as negotiating IVR menus and hard to navigate websites – and matched them with the ‘mirror’ image of those challenges – how to fix them and beyond, to provide a very different kind of experience.
Globe recognized that self-care channels are good for customers and service providers alike, if they’re done right. This is because customers like the speed, convenience and 24-hour availability, while service providers want to improve customer service, but need to lower the associated costs.
Championing the customer cause
Globe proposed this difficult issue as the basis for an accelerated, collaborative R&D project within TM Forum’s Catalyst program. It was accepted and Globe became the champion of a project “to explore the use of cognitive computing or artificial intelligence technologies, such as natural language processing and machine learning, to create the next generation of customer care services.”
Globe was supported by participants Amdocs, Huawei and Infosys. One of the key parts of this initiative was the creation of a cognitive digital agent that could answer questions on various channels, including IVR, and only funnel the more complex queries or those that cannot be resolved automatically to human beings.
Clearly a key requirement was that it understands natural human voices and combines artificial intelligence, machine learning and the latest customer relationship management (CRM) techniques to produce answers. Not only does this streamline how each query from a customer is handled, but the machine learning element means that the more queries it deals with, the greater its knowledge base. Further, the agent was intended to make interchanges, which customers used to find exceedingly irritating, fun as well as efficient – no mean feat.
The toughest aspect
This is easy to say, but the machine learning element, which was the major focus, was also the most difficult. This is new territory and questions like, “What are the best ways to train a cognitive digital agent so that it can genuinely support and improve the customer experience?” are entirely new territory.
The Catalyst was also intended to understand how artificial intelligence and machine learning can augment – rather than replace – the human workforce.
Using TM Forum’s assets
The team used TM Forum’s Open APIs to integrate all the elements of the project and drew on the Forum’s pioneering omnichannel work, which has been tried, tested and progressed in a whole series of Catalysts. The single cognitive computing platform, linked to the CRM, is a big move towards breaking down silos and dramatically speeding up implementations.
Nik Willetts, Deputy CEO, TM Forum, commented, “This Catalyst has the potential to reduce costs by between 10 and 40 percent, and Globe intends to implement it over the coming months” on announcing The Cognitive Digital Agent has won the best-in-show award* (out of the ten Catalyst demonstrations) at the recent TM Forum Live! Asia event.
He added, “This winner showcased the business outcomes exceptionally well, and the inventive use of machine learning and artificial intelligence”.
We are the champions
What did Globe make of its first Catalyst experience and the outcome? We asked Vincent Seet, Head of Enterprise Architecture, Globe Telecom. He said, "It has been an exhilarating journey over the last four months… [Once the value proposition was understood] we had the sponsorships from the senior leaders in our organizations and the project team went out full force to deliver the vision of the Catalyst.”
He added, “The project team has certainly learned a lot from this Catalyst project. It has become clear to us that the Catalyst program is a very effective platform to encourage innovation and solve problems faced by the community. The collaborative model is something that service providers would not be able to replicate on their own without always having commercial obligations.
“Apart from that, TM Forum has taken care of the contractual aspects in the collaboration, which allowed the champions and participants to truly focus on solving the problem at hand. The TM Forum community should really leverage this platform to its fullest potential."
What now?
Looking to the next steps, Richard Im, Director of Carrier Software, Huawei, said, “Going forward it’s more [about how] omnichannel experiences and different kinds of analytics for understanding customers’ behavior could connect with what we’re doing. That will give us a lot of information to see what the business can do and acting on it to monetize, but also to retain customers.
“It could be used to develop different business models with other industries in future, combining those elements together. So, this is a great starting point and from here we could move into all sorts of areas, including collaborating with other Catalyst projects – we have a vision of getting together with other Catalysts and coming out with new, innovative ways to work together.”
Watch the video interview with Vincent and Richard here and stay tuned to Inform to see what happens next.
* The judging panel judging panel comprised Jenny Huang, Lead of OSS/BSS Standards Strategy Group, AT&T; David Pleasance, Chairman, TM Forum; Michael Lawrey, Board Member, TM Forum; and Nik Willetts, Deputy CEO, TM Forum. Check out the other winners here.
