According to a report published by consulting group Opimas, the financial services sector was projected to have spent US$1.5 billion on artificial intelligence (AI) in 2018, rising to US$2.8 billion by 2021. Much has been made of the expected rise of robotics in the 21st century, but Amazon, Google et al have pointed the way to a future in which AI applications consume and interpret data to better meet human needs.
In the financial services sector, the challenge is to follow that lead, delivering tailored outcomes that enrich the lives of customers. Here are some ways banks, fintechs and financial institutions can harness AI for success in their digitization journey.
1. Make clean, high-quality data a key priority
It was not so long ago when companies would look to generate new business by sending marketing messages to prospective clients by post. Direct or “junk” mail – sending a leaflet or letter to anyone for whom the firm had a job title, street address or postcode – was considered unsophisticated, and was often derided as a “machine gun” approach.
Although some still use this method, others sought to exploit the greater reach, creativity and traceability of digital marketing when trying to get a message across or identify demand. Some methods still rely on relatively little data, such as an email address, while others combine many, from past purchases to browsing history. Nevertheless, too many of the finance sector’s current client marketing and communication efforts continue to employ the machine gun approach.
Greater precision in finding the target comes only with better knowledge, or, more specifically, data. Before you can talk AI or machine learning, one must discuss data. In our increasingly digitized knowledge economy, we are generating, capturing and analyzing data at volumes and speeds that are accelerating exponentially.
Data is value for the modern organization. AI is reserved for companies who will figure out how to use quality data to build a stronger business for the future. That said, in order to derive value from data to build a better organization, you’ve got to have the right systems, the right employees, the right processes and possibly most importantly, the right attitude in place.
Before starting any kind of machine learning or artificial intelligence development programme, it's the role of management to make sure the quality of their data is ready for it. Through access to high-quality data, AI programmes can ensure messages to and communications with customers hit the correct target time after time.
The more an AI programme knows about a client’s preferences and priorities, the better it can meet and even anticipate future needs. Already, AI-driven apps are proposing courses of action and making recommendations for users’ consent via a click of a mouse or swipe of a screen. Before long, these interfaces will fall away, leaving just an AI-enabled conversation between the customer and the financial service provider.
2. Focus on the client experience
Banks currently interact with customers across a variety of channels, from branch to phone to web to app. Inevitably, a lot of potentially useful data slips through the gaps that could otherwise help draw an increasingly detailed and accurate client profile.
AI has existed in banks’ back offices for some time, but it is fast emerging into new roles, and could well become a core competence in the near future, becoming critical in efforts to optimize customer interactions, ensuring they deepen and broaden over time to sustain valued, trusted relationships.
Already chatbots are making great strides in making the process of interaction more natural for the customer and more useful for the provider in terms of understanding and gauging future needs. As with the virtual assistants that are almost invisibly managing our homes, banks’ AI interfaces could also change the user experience landscape as they reach maximum utility.
It may be a bigger challenge for banks to join tech giants at the “conductor” level, where virtual assistants orchestrate an array of services and capabilities, bringing them to the user at the precise point of need. But even if they are not yet fully deployed in financial services, the tools, skills and capabilities that will help banks begin their own AI journey are already fast emerging and available.
3. Have a clear roadmap and don’t be afraid to start small
One might say that banks are moving into the “curation” phase, with AI programmes making recommendations based on past experience, in the same way Spotify might recommend a song or Netflix a TV show. It may not be too long before a virtual wealth advisor tells a client that the stock he declined to buy a few weeks ago has dipped in price, potentially making it an even stronger opportunity than before.
Today, this interaction might be effected via an email, a text alert or a chatbot, but in the long term, it might be part of a personalized, one-on-one dialogue, albeit driven by AI. Saxo uses AI and machine learning at every stage of the customer journey from client acquisition through to trading services and retention.
It’s a long journey from direct mail to “anywhere, anytime” conversations between bank and customers, but AI has the potential to define the user experience in financial services such as wealth management just as much as in other customer-facing businesses. The app, just as much as the teller, will cease to be the interface or the voice of the bank.
As the knowledge economy matures, financial service providers should perhaps consider where they are on that journey and where they want to get to. For many, it may be a relatively small step to the “curator” level, starting with a focus on the desired client interaction, rather than the technology. Indeed, small steps often give the best chance of new initiatives gaining momentum over time.
Chris Truce is the head of fintech, Saxo Bank.