There has been a lot of interesting discussion on how bots, and especially chatbots, could redefine the future of customer service and customer experience. The rise of bots has been exceptionally fast, and Facebook played a significant role in the chatbot revolution when they announced in April 2016 that Facebook Messenger platform supports chatbots – the number of chatbots on Facebook Messenger has been growing fast.
We have been living in the world of (chat)bots for a couple of years now, and Facebook’s adoption of chatbots has been a significant landmark in the more general adoption of bots. There has already been some evidence that chatbots are gaining legitimacy amongst consumers. According to Aspect Software Research’s report published last year, “44% said that if a company could get the experience right, they would prefer to use a chatbot or automated experience for CRM.”
This discussion has been, of course, promoted mainly by those individuals and companies that want to push forward their agenda but thanks to vast array of opinions out there, it’s easier to make informed judgements about the potential opportunities and challenges when working with (chat)bots. As Mats Kyyrö, Head of Discovery team in OP Lab, highlights in a recent article, there are various resources and extensive guides available to actually experiment with chatbots.
It seems that there are already different areas of application for bots and chatbots.  We’ll see how many more use cases will be discovered in the future but something that I’ve noted that several players in the financial services industry all over the world have at least announced to give a try. This is a bit surprising as the financial services sector is generally viewed as extremely conservative and hard to change, but as many recent examples have demonstrated – maybe one should follow the phrase: “Past performance is not indicative of future results.”
As in the case of retail, banks, brokers and credit unions will certainly face their own issues and challenges with chatbots. There are several issues related to strategy, operating model and execution that might hinder the transformation of customer service. In banks and other financial services companies, most of the internal issues are indeed related to technology, processes, and organizational capabilities. On the other hand, regulation and compliance, and customer behavior are having an effect on the implementation of automation as well. For example, a recent study conducted by Accenture shows that a significant part of older clients, particularly in terms of using digital wealth management services and preferred communication channels, are more willing to use non-digital alternatives.
Although there are numerous external and internal barriers to implementing, for example, chatbots and other forms of bots, things are changing, albeit slowly. It’s evident that there are still many things that remain to be figured out, and maybe these uncertainties contribute to the hesitancy and reluctance to push chatbots forward more widely. It seems that banking customers are still hesitant to adopt mobile and digital services for various reasons. As McKinsey’s Henk Broeders and Somesh Khanna point out in their 2015 article that, “analysis suggests that digital laggards could see up to 35 percent of net profit eroded, while winners may realize a profit upside of 40 percent or more”.
Even though chatbots do not constitute a sustainable competitive advantage over rivalry for the incumbent financial services industry, chatbots offer a multitude of options for the incumbents to push forward transformative agenda in the digital space. This doesn’t mean that chatbots on their own would change everything, and chatbots should not be treated a substitute for a more throughout review of the whole service delivery system.
The most important thing to note from the contemporary consumer behavior research is that the needs of the marketplace and the expectations of the younger generations, as well as their economic situations, are quite different from what earlier generations faces. As C. K. Prahalad and Venkat Ramaswamy pointed out already in their 2004 book, The Future of Competition: Co-Creating Unique Value With Customers, that today’s consumers are more informed, networked, and active than ever before, and the effects of these changes in the consumer’s role are further amplified by the convergence of technologies and industries. So it’s not just about banking or financial services anymore but rather cross-industry linkages and dissipation of boundaries. With these general issues in mind, chatbots, bots, artificial intelligence, and machine learning are having additional effects on the extent of co-creation.
As Jim Marous highlights, “80% of financial institutions globally view chatbots as an opportunity. Only 16% view chatbots as both a threat and opportunity, with only 2% believing chatbots to be a threat.”
Are chatbots an opportunity, threat or both for the financial services industry?
It has been somewhat surprising that incumbent Nordic financial institutions have not yet been pushing forward publicly available chatbots or really any kind of bots. Last year it was announced that SEB, one of the biggest Nordic financial groups, uses IPSoft’s Amelia for customer service and has set up its own center of excellence to push the development of Amelia forward. Just couple days ago it was disclosed that Nordea Life and Pensions in Norway has launched Nova, a new customer service chatbot.  As Nordea’s Mattias Fras, Group Head of Robotics Strategy and Innovation, point out in an article published in Finextra, “‘Virtual colleagues’ are already at work in several areas of the business” (so this is not the first virtual assistant that they have deployed at Nordea). I have discussed the variety of OP Financial Group’s chatbot experiments earlier.
There are various kinds of exciting financial services chatbots available as today. The key message is that instead of doing all the research and information gathering by yourself, booking a meeting to a bank branch office, or trying to remind yourself if this or that piece of information is relevant, chatbots can actually be used to automatize almost the whole end-to-end process (probably not for now but in the near future even the most complex personal financial transactions, e.g. estate planning and tax optimization, can probably be carried out with the help of a specialized bot). As Gary Vaynerchuk has insightfully pointed out about a year ago, “Being able to buy and sell back time is what some of the most successful startups over the last 3 years have done. They capitalize on our craving to get back the time we have to spend on the things we don’t want to do.” It’s easy to see that fintechs and other startups might be much more eager to start from chatbot-only service model, and therefore it’s essential for incumbents to understand why chatbots exist, what they do, and how they could be utilized.
It’s clear that the service is not improved just by the chatbot alone, but there is a strong prima facie case in support of giving them a try – a client can hopefully enjoy convenient, smooth and smart service from a banking chatbot, and the service providers can actually focus on accomplishing things that are perceived more valuable. Chatbots and other types of bots, at least for me, represent an actual possibility to liberate customers, drive down costs and build more genuine engagement opportunities throughout the service delivery system.
Maybe bots aren’t so scary after all
According to different surveys conducted by Personetics and Accenture, 70% of consumers welcome AI-powered financial services and more than 75% of financial services companies perceive chatbots as an opportunity. As I pointed out at the end of the earlier section, there are various benefits of adopting new client-facing interactive technologies for both customers and businesses. Chatbots are not toys but rather they should be put into use, i.e. learn as much as you can about the various bot technologies available, figure out analytically if there is any potential in them in your particular situation, and most importantly, draw a technology roadmap and run a pilot. Accenture’s recent paper, Chatbots in Customer Service, is an excellent starting point for your journey.
Contemporary financial services landscape is characterized by extreme data-intensity, and as every transaction, whether drawing a cash from ATM, initiating a payment, or applying for a mortgage, creates some sort of data, there are multiple opportunities for carrying out more advanced big data analytics to figure out what’s really going on (and business and data analytics capabilities are more important than ever). Chatbots, on the other hand, should go side by side with any big data project. Chatbots still mostly rely on predetermined pattern of interaction, and because most retail banking operations and processes can be defined with some effort, chatbots and other forms of bots (combined with analytical insights) can actually be the right way to carry out transactional end-to-end flows (not the whole customer decision journey, of course, but some parts of it). As Christoffer O. Hernæs argues, “Banks are in possession of vast amount of user data that is invaluable for unsupervised machine learning as well as intangible know-how through years of experience for supervised machine learning.”
Jan Dinkelspiel, while he was the Chief Innovation Officer at Nordnet and my direct boss, emphasized the importance of mobile-first banking strategy. This idea is strongly supported by the evidence available as various kind of mobile devices, and especially smartphones can be found almost everywhere. As Antero Kivi points out in the abstract of his doctoral disseration, “the diffusion of mobile Internet services as systemic technological innovations depends on a cluster of separate but interrelated technology components that diffuse interdependently due to both demand-driven adoption and supply-driven dissemination.” From this, it can be further concluded, that the diffusion of the Internet is correlated with the usage of (new) mobile devices.
The number of smartphone users worldwide will be almost three billion by the end of 2018, and top 5 mobile messaging apps had over four billion active users as of January 2017. Both WhatsApp and Facebook Messenger have over 1 billion active users, and new messaging services such as QQ Mobile, WeChat, etc. have a large active user base. The volume of messaging processed by these services on a daily basis is humongous, and as most people are already aware of the basics, the potential for utilizing messaging platforms for delivering various kinds of services, with the help of bots or not, is tremendous. As every industry imaginable is always pushing out new mobile solutions, people are regularly and conveniently using their mobile devices for multiple purposes and as technology continues to change, chatbots, if done right, are probably one of the smartest ways of reducing the demand for human (customer) service. Wouldn’t it be great to access your wealth manager whenever you want, wherever you want?
The financial services industry is often represented as risk-averse, traditional, conservative, defensive, brand-sensitive and wanting to protect the status quo by proclaiming the importance of trust and relationships, there have been many strong examples of doing things differently and pushing the limits of the industry. Although it’s still far-off that every bank has their own Facebook Messenger app (and there are grave security concerns), there are many quite promising examples of going towards this direction – take a look at DBS (Asia), ATB Financial (US), Wells Fargo (US), imaginBank (Spain; part of CaixaBank), BBVA (Spain), and TransferWise.
Banks need to grasp the importance of Facebook Messenger and other messaging services alike if they truly want to be part of their customer’s everyday life. “The bots for messenger platform represent a unprecedented opportunity for incumbents to increase customer engagement,” Christoffer O. Hernæs points out. Banks, insurers, and other financial services companies should not stay passive in the era of the customer, and therefore it’s crucial to figure out how chatbots, virtual assistants and other forms of new emerging interactive technology can be harnessed for the good of the whole industry.
The critical questions a financial services companies must ask is how to make excellent self-service as part of their value proposition and take closer look at the whole customer journey instead of locally optimizing single touchpoints. Touchpoints are irrelevant if the service itself doesn’t live up to the promise.
There is still a lot of hype around chatbots, artificial intelligence, and machine learning at the moment. Even the skeptics should embrace themselves, and find out, what chatbots really are and if there is a possibility to try bots out in client-facing situations. First, try to find interesting use cases and demonstrations of chatbots in action to understand their opportunities, move on to test chatbots in a non-production environment as a proof of concept, and if chatbot is found to be of value to the service delivery system, see suitable way to pilot the solution. If everything goes as planned, and chatbot lives up to the promise, maybe it’s time to move into full production mode. Everything doesn’t happen at once, and even failures can provide valuable learnings for the whole organization.
With all this in mind, remember what William G. Pollard once said: “Without change, there is no innovation, creativity, or incentive for improvement. Those who initiate change will have a better opportunity to manage the change that is inevitable.”
 For further reference, see the following sources:
CB Insights (2016, Sep 7.). “51 Corporate Chatbots Across Industries Including Travel, Media, Retail, And Insurance.” CB Insights.
Faggella, D. (2016, Nov 15). “7 Chatbot Use Cases That Actually Work.” TechEmergence.
Halzack, S. (2017, Apr 18). “Retailers look past apps to the next frontier of digital shopping: Chatbots.” The Washington Post.
Itsquiz (2017, Feb 27). “Chatbots for retail and e-commerce.” The Mission.
Itsquiz (2017, Mar 6). “Chatbots for retail and e-commerce 2.” The Mission.
Itsquiz (2017, Mar 13). “Chatbots for Retail and E-commerce — Part Three.” Chatbots Magazine.
Nield, D. (2017, Jan 16). “Why Retail Brands Should Start Making Chat Bots in 2017 + Infographics.” Stanfy.
Quoc, M. (2016, Jun 1). “11 Examples of Conversational Commerce and Chatbots.” Chatbots Magazine.
 Nova is based on Boost AI’s cloud-based chatbot platform. You’ll notice that Nova is built on Boost AI’s “James,” and this is probably a significant case for this startup. I am not familiar with Boost AI so I can’t really say how well-known they are, but at their website, they have a couple of references from the financial services industry, e.g. Norwegian SpareBank and BN Bank.