First Comdata Digital Practitioners Meet Up at the Vatican City
Technological advances in customer care over the coming decade will be an incremental revolution, but no less radical for that. The era of labyrinthine multiple-option recorded messages is giving way to a new age of friendly chatbots that use predictive analytics and artificial intelligence to deal more effectively with incoming queries and “augmented” agents with dynamic mathematical models at their fingertips. For some consumers, especially the computer-savvy, digital natives of the millennial generation, the advent of ever-smarter technologies may be almost imperceptible but, for companies, the effect on the bottom line promises to be striking.
“This is not about Big Bang, this is about a small robot doing credit card acceptance or address changes, this is how it will happen,” said Ronen Melnik, Head of Comdata Digital Division, at a gathering organized by Comdata in Rome with European decision-makers from leading companies in the Telco, Energy, Banking, Retail and Technology sectors.
In the hallowed setting of Vatican City, steeped in two thousand years of history, a window was opened onto the future that will be unfolding in relations with customers over the next five to ten years. “Each sector and market moves at a different pace in different places,” said Ronen. “After the hype settles over some of the new innovative technologies, people will start to think ‘OK how do I use this to the best advantage?’”
Comdata Digital is in pole position to observe from close-up this evolution in how companies engage with their customers. On behalf of its clients, its teams conduct millions of telephone conversations with consumers every month about a dazzling array of issues, from automotive needs to telecommunications bills, entertainment, daily transport, leisure plans, shopping and difficulties with a whole range of technological appliances.
RECALLS, PREDICTIVE ANALYTICS AND BIG DATA
Increasingly, interchanges between companies and customers are being facilitated by developments such as predictive analytics and operating systems based on Artificial Intelligence that crunch terabytes of big data in the blink of an eye. One key area where such developments have the potential to make major improvements in customer service and boost profits is the ‘recall’, the moment when a customer gets in touch with a call centre or directly with a company to lodge a complaint or simply raise a problem.
“What if we could predict these recalls better and faster and react within that window of just one hour? Real-time interventions in place of the current lengthy delays,” Nicola Benedetto, Head of Services, Processes and Innovation Engineering at Comdata, asked around 30 of Comdata Digital’s key clients from all over Europe. Predictive analytics and Artificial Intelligence now make this kind of speedy intervention possible. “We need to react as fast as possible to the first call and although the customer relations process is very complex, there are elements that we can identify that can completely change the paradigm,” said Benedetto. “In terms of human behaviour, we may not be able to influence but we can predict.”
MACHINE LEARNING AND PREDICTIVE ALGORITHM
A wealth of data can be extracted on the many variables along the entire chain of the customer journey that can make accurate prediction possible. “To do this, we use machine learning. For example, we analysed more than one million, two hundred thousand incoming calls at our call centre,” explained Adriana Mina, Data Analysis Manager – Big Data Analytics, Comdata. “We looked at more than 160 variables and, of these, 81 were recurring features that were always present when dealing with a ‘recaller’ customer. From this we developed a predictive algorithm that made it possible to understand in near real-time whether we were dealing with a recaller or a non-recaller customer.”
MACHINE LEARNING, AI AND CUSTOMER SATISFACTION
Using a dynamic mathematical model to identify the type of incoming calls has clear benefits in swiftly and efficiently directing the customer to the right agent or specialist. The analysis also revealed some other interesting aspects. For example, the length of time spent with an agent is more important in terms of customer satisfaction than the seniority of the agent. The model applied at the Comdata call centre achieved 78% accuracy rates but, through machine learning and Artificial Intelligence, the accuracy of the predictions increases exponentially. Moreover by adding more variables - credit checks, personal details such as age range – the hit rate goes up still further. So a company can prioritize the customers that require intervention and recall rates can be reduced dramatically.
This level of predictability opens up possibilities of automation and a much more widespread use of robotics, already embedded in the CX world. “We’ve been employing technologies like automated conversation for 10 years. We didn’t call it a chatbot, we called it Automatic Scheduling System (ASS),” said Ronen.
IT'S ALL ABOUT INTERACTION
Consumers though, still prefer interact with humans and so the pace of change is limited to the speed of adoption. “Unlike the word transformation, which implies big boom, things in the real world develop at a much more gradual pace. One reason is that people, the consumers, adopt slowly. Even though you have an app and you can check your account balance, people call a contact centre to ask how much money they have. This is because it’s a habit and habits die hard, so change is slow.” In large companies, the pace can be even slower, Ronen pointed out, because, while the need to do things differently is generally acknowledged, change has to be digested by an entire organisation that has an annual cycle, a fixed budget and multiple departments that need to form a view. “The third reason is that there are a lot of conflicting messages, so what is it that I should be doing? Customer care is very important, everybody wants to care more for the customers in the digital environment yes, but we also want to automate and we also want artificial intelligence, and what is that good for?”
"GREETER 3.0": A MIX OF HUMANS AND ROBOTS
Based on Comdata’s experience, the answer to that question is a judicious combination of a new type of highly-trained “augmented” super-agent, acting on real-time data, and full automation. This formula has been dubbed “Greeter 3.0”, the skilful deployment of a mix of humans and robots that slashes costs and boosts customer satisfaction. It may be some time before the all-inclusive, “broad spectrum” robots like Alexa and Siri will become ubiquitous in Customer Care, ready to ask customers ‘What can I help you with today?’ and provide an infinite range of answers, but “task-specific bots” will increasingly play an important role in the customer journey.
SANOFI: REDUCE THE INTERACTION BY INCREASING CUSTOMER RELATIONS
This kind of targeted automation was a key feature of Comdata’s service for the Paris-based pharmaceutical giant, Sanofi, a leading global supplier of medicines to pharmacists and hospitals in more than 100 countries. The company came to Comdata with an urgent problem. Its pharmacists’ platform was facing an upsurge in interactions, 90% of which were conducted by phone, with soaring costs as a consequence. Many of these were simply requests for copies of invoices, interactions that needlessly tied up agents yet produced few sales. While costs were rising dramatically, the value the company was able to extract from this consumer engagement was vanishingly small.
The threefold brief was to simplify the pharmacists’ customer journey, develop interaction with added value in terms of up-selling and become the pharmacists’ preferred supplier. Comdata conducted a wide-ranging survey with pharmacists to identify the tasks that could be automated. It developed an integrated IT solution to increase the agents’ efficiency and produced a management programme, including training, for the pharmacists and agents. “Sanofi’s goal was to be imprinted in the mind of the pharmacists as the preferred partner to develop their business,” explained Frédéric Donati, General Manager of Comdata France. “We achieved that and more.”
SANOFFICINE: THE NEW PORTAL WHERE INTERACTIONS ARE FULLY AUTOMATED
The result was Sanofficine, a new portal for the pharmacists where many of the interactions are fully automated. It now deals with 50,000 contacts a year, mostly concerning invoices, credit notes and follow-ups on deliveries, freeing up agents to engage with customers in a way that produces high value. The results were impressive. Customer satisfaction went up by five points to more than 90%, the Net Promoter Score increased by nine points and the resolution of problems at the first-call stage rose by 10 per cent. Sanofi was so delighted with the results, it gave the project an award as its Best Business Innovation project. Now the programme is being rolled out for 3,000 hospitals.
“Caring for customers is not a process, it’s an attitude,” concluded Ronen. “I think the technologies and the touch points are pretty much all there and now it’s just a question of how to use them and how often.”