Doron Youngerwood, a product marketing manager at Amdocs, sees communications service providers (CSPs) engaging with machine learning and artificial intelligence (AI) with an initial goal of improving customer experience. However, this is the foundation work for automated, self-healing services, he tells VanillaPlus managing editor, George Malim
George Malim: What stage are communications at in their deployments of AI?
Doron Youngerwood: Service providers are in a pilot stage and are at different stages according to their maturity level. They are trying AI to address certain use cases and specific challenges to prove the value and expand deployment into other areas. For example, Vodafone Ireland is using analytics and machine learning to drive penetration of its mobile app and using these technologies within the app to drive promotional offer acceptance. Vodafone Ireland wanted to get closer and be more relevant to its customers and is using machine learning to calculate purchasing propensity scores among its customers.
The results are promising. It has achieved a 15% increase in net conversion rates of mobile apps users.
GM: How should CSPs approach their introduction of AI-related technologies?
DY: There’s a two-pronged approach. There’s a medium-to-long term view if a service provider is moving in the 5G direction because it will have to have AI to operate effectively. However, the short term view requires some of the capabilities that are available today to reduce churn and improve customer loyalty. Customer experience-related activities are where the focus is today but in the coming year we will see deep learning used in the network domain.
GM: Are CSPs focusing on the long term possibilities or looking for fast gains today?
DY: It’s the second. Where AI is focusing peoples’ minds is where there’s immediate value and immediate return on investment. They need to address things today rather than things that will happen in the future.
On the other hand, people are looking at a more holistic approach to AI because it will be so central to operating their businesses. AI is critical for service providers. They have huge complexity and manage vast volumes of data so they know there’s a return in there but we have no idea how to extract it.
GM: Do you envisage CSPs becoming the enablers, managers and operators of AI hubs that enable future business cases across a more widely distributed ecosystem of collaborators?
DY: Yes, I think that is happening. CSPs are developing 360-degree customer profiles and the technology has been around for some time but, with the advent of AI, it has become more relevant.
CSPs have between 50 and 60 systems that contain customer data so bringing that together and correlating it and putting machine learning key performance indicators on top of that to identify customers’ intent is bringing value. This is where AI really comes into the mix.
With service providers expanding into media, financial services, smart homes and other areas a single vision of customers becomes more and more important. It’s the backbone to any AI strategy. It’s having the 360-degree view that enables the service provider to identify which mobile customers want a new pay TV package, for example.
GM: What does Amdocs bring to this?
DY: We have several key modules and I would emphasise our data integration capability. We manage some of the largest and most complex data environments in the world so we have the data products and our logical data model which are fundamental aspects. Being able to collect data in real time and match it to customer identities and developing machine learning models on top of that means we can orchestrate the differences in customer data. It’s not just about campaign management but customer lifecycle management.
In the last six or seven months a number of service providers have used AI in the marketing domain but, more recently, customer service directors are battling to free up contact centre resources. AI will help to augment those contact centre agents and enable them to focus on high level issues.
Billing-related issues, for example, can be solved by chatbots and in the future AI will be applied to data to provide proactive care before the customer even picks up the phone.
This is how CSPs can make use of customer data to automatically identify patterns that lead to calls to the call centre. Ultimately this will enable predictive resolution of issues and even self-healing.
