by Thor Olavsrud

Nokia uses analytics, machine learning to help mobile providers

News Analysis
Nov 23, 2016
AnalyticsBig DataCarriers

As mobile competition heats up, Nokia is betting that advanced data analytics and machine learning will help wireless carriers optimize their customer service.

nokia primary3
Credit: Thinkstock

Wireless carrier competition in the U.S. is white hot — analysts increasingly see signs of wireless market saturation, meaning that growth is most likely going to come from competitors. Of the four major U.S. carriers, T-Mobile USA has led the way, slashing prices, killing the two-year contract and daring its competitors to follow suit.

As competition intensifies, Finnish mobile technology provider Nokia believes customer service will emerge as an even more important key differentiator, and analytics and machine learning will take customer service to the next level in the U.S. and around the globe.

In 2011, when AT&T announced its intention to acquire T-Mobile USA from Deutsche Telekom, it looked like the wireless market in the U.S. was well on its way to becoming a duopoly. Verizon Wireless and AT&T already held the lion’s share of customers between them, and they both held licenses to the majority of wireless spectrum too. But then the Antitrust Division of the U.S. Department of Justice blocked the proposed $39 billion acquisition.

T-Mobile USA, then by far the smallest of the big four, had to change its tack. And it did. It abandoned contracts that locked consumers in, offered to pay the early termination fees of customers who left competitors to sign up, introduced lease plans under which customers could have access to the latest phones as soon as they launched and more. T-Mobile still has the smallest network footprint of the big four, but it has surpassed Sprint to become the third-largest mobile provider in the U.S. — though AT&T, the second-largest, still boasts roughly twice as many subscribers.

Enter Nokia and analytics

Still, the game has changed and the wireless carriers are seeking every edge to get a leg up on their competitors. Enter Nokia’s Motive Customer eXperience Solutions (CXS) software portfolio, intended to help communications service providers improve customer experiences. Last week, Nokia updated two components of CXS — Motive Service Management Platform (SMP) 7.0 and Motive Care Analytics (CAL) 2.0 — with machine learning algorithms developed by Nokia Bell Labs. With machine learning, Nokia believes its customers will be able to set a new standard for proactive care in the industry, dramatically improving the detection, troubleshooting and resolution of subscriber issues.

“If you’re going to do customer care, you need analytics,” says Rich Crowe, head of marketing for Customer and Network Operations, Nokia. “You need analytics on what’s happening in the call center to improve the call center.”

Motive SMP 7.0 features Dynamic Intelligent Workflows, a new self-optimizing system that determines the ideal sequence of tasks that deliver the highest probability of resolving billing, subscription and network service issues in the shortest amount of time. It analyzes data from previous workflow executions, the network, customer premises equipment and trouble tickets, helping service providers find the optimal remediation to issues, whether customers are contacting the help desk or using self-care apps.

Motive SMP 7.0, for instance, can detect which remediation you took through a self-care app before getting frustrated and calling customer care. As a result, Crowe says, customer care agents won’t ask you to repeat remediation steps that have already failed.

It works in conjunction with Motive CAL 2.0, which automatically correlates customer help desk calls with self-care actions with network, service and third-party application topologies to identify call anomalies, like unusual patterns in help desk calls that indicate the location of customer-impacting network and service issues. Once it identifies such anomalies, it initiates actions through Motive SMP to resolve service disruptions and other issues before they become widespread problems.

“It allows us to quickly spot what we call anomalies coming into customer care,” Crowe says. “What is a normal flow of requests for a given day or a given service? If it spots something outside of the norm, it then is able to correlate a small series of anomalous calls and determine that we may have a larger issue. Then it can engage the network service operations center to resolve larger problems before a flood of calls or care requests come into the call center.”

In so doing, Crowe says the system can deflect up to 85 percent of call center calls by providing messaging about outages and steps the carrier is taking to resolve issues, which will improve customer satisfaction and reduce churn.

“If it does get to a care agent, the care agent is then also informed of service disruptions so that they can then handle the service request in the context of that larger disruption,” Crowe says. “The service agents now are all acting with the same information that the call anomaly system has detected a larger disruption.”

Cutting help desk and truck times

In all, Nokia estimates SMP 7.0 and CAL 2.0 will help service providers reduce average help desk handling times by five to 15 percent and eliminate inappropriate truck rolls related to network outages by as much as 90 percent.

“Service disruptions are often hard to identify because they happen in the access network, on customer equipment or on customers’ devices,” Bhaskar Gorti, president of Applications and Analytics at Nokia, said in a statement last week. Traditional customer care may only address a small part of a larger problem and the time-consuming, step-by-step troubleshooting process can lead to customer frustration and the risk of lost business. By providing the earliest possible detection of network issues and streamlining help desk and self-care interactions, these new Nokia solutions reduce IT and care costs, and result in happier, more loyal customers.”

“Reducing the complexity around customer care, providing a personalized experience and introducing self-service are key to maximizing customer satisfaction and improving Net Promoter Scores,” adds Sheryl Kingstone, research director, Business Applications, at 451 Research. “By embedding machine learning into customer care offerings, operators can provide relevant and engaging customer journeys across new channels of interactions, which are integral to success in today’s digital world.”