Are Credit Unions Up to the Chatbot Challenge?

“Human touch” has long been a pillar of the credit union industry, but the rise of robot-guided chatbots could soon redefine what it means for members to engage with their credit unions – and even what kind of engagement options credit unions provide in the first place.

Few credit unions use chatbots right now, but data suggested that the tide is changing in favor of credit union chatbots. A recent Salesforce survey of 3,500 customer service professionals, for example, found that 53% of service organizations expect to use chatbots within 18 months, which is a 136% growth rate.

Whether credit union members will come along for the ride is still up in the air. A recent survey by chatbot developer Drift found that people seem to prefer chatbots when they’re in the right place at the right time – 86%, for instance, would rather use a chatbot than fill out a form on a website. On the flip side, a different survey by enterprise and outsourcing solutions company CGS found that only 29% of consumers preferred chat or direct messaging for quick questions, and just 16% preferred it for complex questions. If the consumer is stressed or anxious about something, the preference for chat dropped to just 12%, it noted.

What’s perhaps slightly more predictable is the lucrative impact that chatbots could have in the credit union space. Chatbots have the potential to slash call center costs, for example, and they could even help sell more products or streamline lucrative onboarding and application processes – if credit unions use them effectively, experts said.

Roboseduction

Chatbots are text-based or voice-based conversations between human members and a robot. They’re not the same as “live chat,” where a member trades real-time messages with another human. Many text-based chatbots are visible in the corners of credit union homepages or on credit union mobile apps.

Although they may seem relatively cold or impersonal compared to phone support or even live chat, chatbots have one unavoidable advantage – speed.

A chatbot can handle a call in about half the time it takes for a human-guided live chat to do it, which could translate into major labor cost savings, according to Alan Rosenbaum, EVP at the Portsmouth, N.H.-based credit union chatbot developer SilverCloud. As many as 40% of the questions that flow into a typical call center are routine, chatbot-answerable questions, he noted.

“If I take 10,000 calls a month, and that means for 10,000 calls per month I need five full-time equivalents because an FTE can handle about 2,000 calls per month – if I get rid of 40%, that’s two head count,” he said.

Chatbots are also tools for engagement, Derik Krauss, senior website consultant at the Cottonwood Heights, Utah-based BloomCU, which builds chatbots and does website design for credit unions, said. Chatbots are particularly good at three things: Guiding members through a specific process (such as membership enrollment), recommending products and services, and answering frequently asked questions, he said.

Not all chatbots work the same way, though. Some can only answer simple questions; others can handle more sophisticated, natural human language. Some chatbots lead users through quiz-like questionnaires rather than trying to intuit member-typed sentences. Some have novel features, such as reporting how much the member has spent on certain items.

Chatbots Don’t Cost the Same

“It’s really all over the board,” Krauss said. “There are lots of different campaigns that offer chatbots, and the technology is different.”

One of BloomCU’s technologies runs about $100 a month to get started, according to Krauss. Options from other companies might run about $400 a month for one credit union employee user account or $600 a month for five user accounts, for example. Credit unions often spend far more, Rosenbaum said.

“We have a client that we’re working with who has spent half a million dollars trying to get a much more useful chatbot,” he said. “And we’re not even talking about one where it’s got deep data integrations. We’re talking just sales and support.”

Three Tips for Credit Unions

Chatbots aren’t a plug-and-play affair, both experts warned. Before getting started, credit unions should think about three specific things if they want to get the most out of a chatbot.

1. Avoid bot rot. Somebody has to write the words the chatbot puts in front of members. That person might work for the fintech partner or the person could work for the credit union, but either way, someone needs to be responsible for maintaining the chatbot and ensuring that the information, guidance and content their chatbots provide is always up-to-date and useful. “You really have to set up the content and then constantly improve on that content,” Rosenbaum said, “and then you’ve got to have the technology to support that content and process.” Letting a chatbot “rot” runs the risk of spitting out old or wrong information, leaving members to wonder if anything the credit union is telling them is correct.

2. Guard the integrity of the omnichannel experience. Chatbots can collect and process amazing amounts of data, but that may be a huge headache if they don’t work with other credit union systems. In particular, credit unions need to know what they’ll do if a member stumps the chatbot and needs to talk to a human. “The number one consumer complaint when using any kind of digital support like a chatbot is that, ‘I try in the digital channels, and when it doesn’t work and I go to the human channels, whether it’s a live chat or I call them, I have to repeat my problem all over again,’” Rosenbaum explained. “How are you going to integrate that chatbot into your live chat or into your call queue and tell that person this is what they’re searching for? You’ve got to start building frictionless interfaces between the chatbot and the systems that you’re using to support the human systems.”

3. Figure out how you’re going to measure the ROI. Determining whether an investment in a chatbot was worthwhile starts with defining the chatbot’s purpose, Krauss explained. “What are you trying to accomplish with a chatbot? Are you getting it for the purpose of growing the credit union, [are you] trying to delight your existing members, or are you trying to do that to cut costs for the credit union? If you know what you’re trying to accomplish within, then I think that’s going to do a lot to guide [credit unions] in the right direction.”