Credit: AI Trends
The banking and financial services industry is ripe for technological disruption. Banks are naturally experimenting with AI (artificial intelligence) to automate traditional day-to-day transactions, while insurance companies and financial advisors are chasing alpha by layering AI on top of data to ascertain risk. These institutions in general are expected to save more than $1 trillion by 2030 by using AI.
Enter the digital customer service representative, virtual agent, or advisor. Whatever you choose to call it, they all (at least the good ones) use AI to automate the handling of thousands of basic questions and endless tasks, that companies face every single day. Everyone has to adapt. It simply is no longer financially or competitively feasible to rely solely on human agents.
But is the decision to implement conversational AI all about ROI? Yes, and it should be, but many firms are leveraging this automation to not only reduce associated costs; but also to uncover new revenue streams, and induce a higher level of customer loyalty. In fact, conversational AI platforms are so sophisticated that Juniper Research forecasts the technology has the potential to slash business expenses by as much as $8 billion by 2022.
European companies are moving even more swiftly than Americans in applying artificial intelligence across banking and financial services, due, in part, to a more centralized regulatory environment. Consider that DNB, which one of the largest banks in Europe, recently reported a rate of return of more than 14 times their investment in virtual assistants for their first-contact resolution pathways.
This top-performing European financial institution was also able to reduce customer chat support by 49 percent, while managing more than 10,000 automated conversations every day. When you compare this to first-contact resolution rates for popular conversational AI agents Zendesk Answer Bot and Ipsoft Amelia, which approach 12 percent and 13 percent respectively, it hard to argue with DNB’s results.
Know where and how to start
To really capture the ROI-boosting value of conversational AI, industry players don’t need to know how to design, develop, deploy, and train AI technologies. Digital technology providers, like Boost.ai, already fill the gap between engineers, data scientists, and business consultants, so that these organizations can focus on what they’re good at, instead of becoming conversational AI experts. But businesses do need to know how to prioritize AI in their customer interactions to ensure it is well aligned to their specific needs.
The advanced virtual agents employed by DNB employ natural language processing (NLP) algorithms and new forms of semantic analysis to achieve complex, purpose-driven conversations. They also tap into a multi-level hierarchy that can process thousands of customer interactions simultaneously, while also hooking into existing CRM and ERP systems to create more efficiency.
1. Constant availability
Today, many companies are expanding call center hours to keep up with customers. These customers are getting more demanding and want help exactly when they need it, not when the company decides it’s available.
Those that have successfully implemented conversational AI, like DNB, are already starting to decrease the availability of their human operators as one of many steps towards digital transformation.
In addition to the cost savings associated with reducing the hours human operators are available, organizations are increasing customer satisfaction by supplying an always-available company representative.
2. Instant responses
No one likes to be put on hold, and a long wait time is unlikely to encourage customers to continue a conversation with their bank or insurance broker. If delays happen frequently, it goes without saying that it will lead to a loss of business, from both existing and potential customers. The customer of the future is likely to be even more impatient.
Ironically, despite this ongoing push for quick service, feedback from large firms in the insurance and financial services industries has been remarkable, in that conversational AI might even respond too quickly for some customers. Because an instant response is the default for virtual agents, Santander, like many of our other clients at Boost.ai, have reported that their customers were thrown off by a fast rate of response.
The solution has been to make the virtual assistant appear more “human” by adding a delay of 1-2 seconds to the response time. The change made it seem like the conversational AI was “typing,” and Santander never received another complaint.
3. Proactive abilities
Many people think that introducing a virtual agent has the potential to lower the quality of the service, but that doesn’t consider the clear and growing trend towards self-service. In general, this is a good thing, because the cost of providing service goes down and customers can easily access information. But some might argue that this automated service reduces brand interaction.
Conversational AI can actually provide the best of both worlds. Taking customers back to “the good old days” when they could talk to “their guy” at the bank or credit union. What if you could talk to your customers like that again, one-on-one? What would you say? How much more revenue would that bring?
Unfortunately, these kinds of conversations would cost a fortune today for large enterprises. But AI-powered virtual agents do still allow for personalization of service.
Storebrand, a provider of insurance and pension products, uses Conversation AI technology to automatically reach out to a customer that recently purchased life insurance; to ask if they are happy with their policy, or if they can help them with their pension savings. This has been proven to delight customers and gain more business.
4. Capacity and scalability
Staffing a call center to tackle peak periods is a difficult task, that often leads to frustration on both sides. Most budget chatbot technology can handle small peaks, but it takes true, innate scalability to tackle a growing number of inquiries over time.
This is also true for sudden, unexpected events, something banks, insurance companies, and financial institutions understand all too well. With proper conversational AI, organizations can handle an almost unlimited number of conversations simultaneously. Preparing for such events will satisfy those that depend on customer service on a rainy day, and also have a positive effect on reputation, whether it’s in a crisis situation, or more generally.
Read the source article in MTA Martech Advisor.