Lowell improves customer interactions and delivers better client outcomes with Azure Cognitive Services

Fair, responsible credit management

Lowell believes credit can work better for consumers, for lenders, for everybody. One of the largest credit management services in Europe, Lowell uses debt purchase, responsible collection practices, fair terms, and continuous innovation to deliver better outcomes for its clients—and consumers. The service’s highly trained staff manages debt recovery, risk, and compliance for clients while also providing payment services and financial advice to help consumers manage credit. 

Still, Lowell has to create value in sensitive and stressful situations, so it strives to make every consumer interaction as stress-free—and even enjoyable—as possible. The company has data that suggests up to 40 percent of unpaid invoices are held by people who have the means to pay, and Lowell motivates consumers to pay by offering credit counseling and other services to help them get out from under their debt.

Unnecessary inefficiency

But in Norway, the legal framework for debt collection is complex, so highly educated Lowell staff members spent too much time handling consumer questions, fulfilling routine requests, and looking up payment records. That’s not only frustrating for trained financial counselors, it’s expensive for Lowell and inefficient for everybody.

“Approximately half of our phone interactions are small inquiries,” says Carl Udvang, Product Manager at Lowell. “Our financial counselors had to spend 60 percent of their time on routine communications with consumers rather than sharing their credit expertise to help consumers tackle their money issues.” 

Lowell knew that freeing its staff from dealing with mundane tasks wouldn’t just help them become more engaged in their work—it would also mean that financial counselors could spend more time delivering better experiences for consumers and motivating timely payments.

A more intelligent, intuitive, and useful chatbot

So, to motivate payments, deliver a rewarding work experience, stay compliant in a highly regulated space, and reduce costs, Lowell turned to its longtime technology partner Crayon for help. As the 2019 Microsoft AI and Machine Learning Partner of the Year, Crayon brought its expertise in emerging technologies to bear in helping Lowell deliver an advanced call center automation solution.

Together with Crayon, Lowell built a custom AI-enabled chatbot using Microsoft Bot Framework and Microsoft Azure Bot Service. They used prebuilt machine learning models in Azure Cognitive Services to train the bot so it could interact naturally with consumers, responding to complicated inquiries as well as answering routine questions.

“A lot of bots won’t understand a question and will default to a generic and often unhelpful response,” says Udvang. “We used Azure Cognitive Services to create a more intelligent, intuitive, and useful bot.”

A flexible, efficient architecture

Lowell uses eight on-premises AI containers to run Cognitive Services locally as necessary to meet its clients’ and its own business, compliance, and performance requirements. With the flexibility to deploy and host AI services on-premises, at the edge, or in the cloud, Lowell can respect consumer privacy and safeguard personal data yet still generate high-throughput and low-latency performance. That gives Lowell the power to run advanced features like text recognition, sentiment analysis, and key phrase extraction, which help its bot answer 40 percent of consumer inquiries—and even calculate a payment plan for each consumer.

“By taking advantage of container support in Cognitive Services, we built a bot that safeguards consumer information, analyzes it, and compares it to case studies about defaulted payments to find the solutions that work for each individual,” says Udvang.

When a consumer has an issue that requires human assistance, the automated call center delivers a seamless handover from the bot to a live service agent. “The switch from bot to human advisor is really important,” says Udvang. “With Azure, we transfer conversations from the bot to a service agent without losing the chat log, so the consumer experiences the entire interaction from a single interface.”

Credit that works

The Lowell bot provides an efficient way for consumers to receive advice, establish payment plans, find answers to simple questions, and manage their own credit. Its introduction has seen a 20 percent increase in first-call resolution rates, from 50 to 70 percent. 

Meanwhile, the company’s service agents now focus on more rewarding, high-value credit consultation. This means better experiences and higher motivation to pay for consumers, higher satisfaction and more opportunity for employees, and reduced service costs, faster resolutions, and better outcomes for Lowell clients. 

“We wanted to give our customers the same experiences they get from other companies in other industries,” says Udvang. “Now we can do that and work more efficiently and effectively on our end, too. It’s exactly what we needed.”

Find out more about Lowell on Twitter, Facebook, and LinkedIn.