The bank of the future will be powered by digital services delivery and AI decision-making. As financial institutions (FIs) continue to adopt and deliver more digital service channels, they also open the door to previously untapped data that can be used to fuel actionable AI decision-making capabilities. The result is a better understanding of customers’ quickly evolving needs and the ability to respond with precision, speed and efficiency.
Marc Jones is Chief Technology Officer at Alkami, and leads the Engineering group including software architecture, development, site reliability, and quality assurance teams, advancing the company’s investments in technology talent and expertise for the future. With nearly 25 years of experience in software development, information technology, product development, and cloud platform innovation, Marc shares his insights into how digital services delivery and AI-decision making will power the “Smart Bank” of the future.
Modern smart banking capabilities are impressive, thanks in no small part to trends accelerated by the pandemic. Artificial intelligence/machine learning-enabled technology such as chatbots, voice assistants and facial recognition existed pre-COVID, but were underutilised.
Then social distancing became the law of the land, and people were forced online in droves. Barriers to adoption that might have taken years to dismantle crumbled in months. And those financial institutions that were built on flexible, scalable infrastructure were able to step up to the unprecedented volume without compromising security.
Now that they have seen the personalisation capabilities enabled by AI and ML in other industries, today’s enlightened users keep raising the bar for FIs. In this “new normal” world, they expect no less than insightful, frictionless service firmly rooted in a real-time picture of prior interactions. From where they sit, it shouldn’t matter whether they engage via the local branch, website, or phone.
The need for this type of access may sound obvious. But proactive, multichannel service remains a bridge too far for FIs that have yet to make it their top priority. Progress has been stymied by a complex assortment of core banking systems, external providers, integration challenges and lack of access to critical data. When so many disparate pieces are patched together, service snafus invariably result.
It’s difficult to predict how much lifetime customer value is lost to one botched transaction or poor service interaction—compounded if it’s replicated over a multitude of interactions across an entire user base. FIs that deliver disjointed service squander that precious customer loyalty while missing out on profitable opportunities to offer new products and services at key touchpoints. It’s a scenario that remains all too common.
Smarter Banking Begins with Smarter Infrastructure
For those FIs looking to ensure seamless delivery across all customer touchpoints, they must focus on three key elements that need to be working together in harmony.
The first is a secure, robust digital banking platform capable of integrating a broad range of first- second-and third-party sources. This permits previously siloed data to be aggregated into a coherent whole to derive a full picture of each consumer’s financial wellness. And given that financial wellness is a long-term pursuit, institutions with a longitudinal view of data can more effectively help their consumers with targeted offers that consider this behaviour over time. Of course, with greater data comes a greater responsibility for safeguarding it properly, ethically, and in accordance with the latest compliance regulations, making the security of the digital banking platform a top priority today and even more so in the future.
But AI/ML models absent human intervention may not translate directly to better outcomes. For example, if the system detects an unusual sequence of transactions, when is it better to send an alert versus freeze the user’s account? That’s where the second piece of the puzzle—humans—comes in. A truly smart system requires not only advanced technology but also talented data scientists to evaluate machine learning, incorporate nuance and ensure an effective and ethical response. The Bureau of Labor Statistics estimates growth for computer and information research scientists to be 15 per cent from 2019-2029, much higher than the 4 per cent expected for all occupations. Financial institutions must be ready to compete in a largely seller’s market for highly qualified data scientists that can help these organisations grow.
Finally, the third critical component is AI/ML that is sophisticated enough to analyse petabytes of raw data, identify patterns and unlock insights into customer behaviour. These learnings can then be used to make intelligent predictions and actionable decisions that better guide users at every point on their journey. Doing so requires abandoning legacy batch processes that still exist in many financial institutions. Users operate in real-time and expect institutions to do the same. With the nearest branch now in the palm of one’s hand, financial institutions that leverage real-time insights to inform a 360-degree view of behaviour can seamlessly bridge interactions that occur in the branch, at the ATM, online, and everywhere in between to meet users where they are…right now.
Rome Wasn’t Built in a Day
Making this vision a reality may sound daunting. But FIs trying to stay competitive on systems hamstrung by yesterday’s technology will only get left behind.
Here are a few tips for moving forward in today’s financial landscape:
- Choose priorities aligned with business objectives. Before you invest a single dollar, decide what operational challenges you are trying to solve. Then determine the information you need and the sources you can tap to obtain it.
- Emphasise quality over quantity. Many FIs have customer or member records going back decades, but not all data is useful or actionable—it may be duplicative, conflicting or dated if based on batch processing. The better the data, the more spot-on the recommendations are from your AI/ML models. This, in turn, enables your human resources to direct their efforts to higher-order goals.
- Balance depth and breadth. Multiple integrations are important because better inferences can be drawn from a 360° view of your customer. For example, customer satisfaction may suffer if a phone service rep fails to incorporate information captured via a chat or voice assistant. Start from where you are, even if that’s only one data set, mine it, and then set your sights on greater breadth.
Artificial intelligence has already transformed our financial lives in myriad ways, but today’s state of the art is just the tip of the technology iceberg. The future of the industry belongs to institutions that leverage their vast, largely untapped data stores to deliver increasingly intelligent, efficient, personalised customer service.