ING Direct of Canada has reduced its call center volume by investing in technology that helped it determine what drove many of its customers to pick up the phone.
In 2009, ING started using a voice analytics product by Verint Systems of Melville, N.Y. Verint's product transcribes call center calls, enabling banks to search for key phrases and words and graph almost any trend.
Using this product, ING determined that a lot of its inbound calls were prompted by easily fixed flaws in its website. For example, consumers were not getting enough information about its retirement product online. ING addressed this problem by adding a product comparison chart. It also created a link to the retirement offering from its homepage.
"We wanted to know what the root of their inquiry was, and why they were not acting online," says David Archard, head of analytics and quality assurance for ING Direct Canada.
By making those changes, Archard says ING, which has spent less than $100,000 on Verint's service, immediately reduced call center volume by 5%. Call times were also shortened, Archard says, because consumers had fewer questions about the product.
On average, calls cost between $4 and $6, according to research firm Celent, of Boston.
ING Direct Canada, a unit of ING Group NV of Amsterdam, has 2.2 million customers, 90% of whom interact primarily online and through call centers, which handle about two million calls annually.
ING has since used the Verint system to get more insight into the behavior of its customer service agents so it could fight off attrition. It was able to determine through phone-call analysis that agents were not reacting quickly or decisively enough to cues from dissatisfied customers other than to ask why the customer was thinking of leaving. The bank jumped in by offering more training and coaching to agents on how best to combat customer attrition, and within a couple of months ING was able to increase customer retention attempts to 70% from 30%. That, in turn, led to a 48% increase in actual retention.
Now, Archard says, the bank is using the technology to measure first-call resolution of problems, and to gauge customer sentiment related to marketing campaigns.
"Large financial institutions are recording millions of phone calls, and they want to give some sort of intelligence to those recorded calls without having to listen to each one," says Ryan Hollenbeck, head of marketing at Verint. That might include sorting calls for information that would help determine which customers are the most high-value, routing them to specially trained call center agents when they call in next.
Many vendors offer speech analytics, including NICE Systems, Mattersight and Novantas Solutions. Evolve 24, a unit of Martiz Research, provides similar services for banks' social media channels, which produce voluminous text-based conversations.
"What people are saying in the open on social media is a very valuable stream of information," says Michael Bryars, executive vice president of technology for Evolve 24. "With social media and [text] processing, we can answer the questions the bank does not know to ask."
The next generation of dynamic speech analytics is called propensity modeling, said Darryl Demos, a partner with Novantas of New York. Based on what the customer says about a product or service an agent may offer, the system can prompt agents with different scripts for different offers.
"We call this dialogue guidance," Demos said.
If the elusive goal here is to do more effective one-to-one marketing in an attempt to get customers to buy, the net effect is that banks and other financial services companies are finding gold in data that was previously considered worthless.
"As a [payments] company, we have always been knee deep in data," says Carey Kolaja, senior director of emerging opportunities for PayPal, in a conversation last month with American Banker.
She said an ongoing challenge for PayPal, a unit of eBay Inc. of San Jose, Calif., is to figure out the best ways to use its consumer transaction data to serve both merchants and consumers.
"There has been a drumbeat around analytics and customer personalization," Kolaja says.
