Introduction
The long-term value of artificial intelligence (AI) [1] in wealth management is immense. The good news is that the foundation for creating powerful AI solutions in financial services is in place. The industry has also reached a new level of maturity, as the use of AI in financial services automation has evolved through multiple iterations, ranging from pure black box solutions created in the lab to various bots and RPA solutions (many of which are artificial but not necessarily intelligent) to “smart” systems. So, given the positive environment for AI, why aren’t we seeing more adoption and value creation?
In the Spotlight
There is currently a favorable an alignment of factors conducive to the adoption of artificial intelligence in wealth management. The economics of computing continue to improve allowing for greater processing power at lower cost. The industry is awash in data ranging from customer data to market data to news to social data. While too much data can be problematic, we now have the ID and tagging utilities plus the database technology required to knit together and derive insight from structured and unstructured the data. On the demand side of the equation, one can find receptive, even enthusiastic audiences such as innovation and digital teams at many financial institutions. Additionally, there is no shortage of media coverage, conference sessions or coding boot camps signaling that AI is poised to transform the industry.
Slow—Speed Bump Ahead
This begs the question: if the environment is so rosy, why aren’t more solutions going into production? There are a few likely suspects.
Organizations & Politics
- An innovation group or a digital team whose mission is to generate new ideas outside of the confines of the everyday operating norms often conceives AI projects. These teams can commission third-party proofs of concept (PoCs)—many of which are successful; but adoption has to be handled by the mainline technology and operations teams. Tech & ops may bristle at having a solution foisted upon them or they may even be working on their own solution.
- Is the top of the house prepared to support AI? Senior management may champion outcomes (e.g., increase advisor productivity, improve customer service scores, etc.), but are they prepared to get behind the associated technology and operations transformation?
User Expectations
- Expectations are often not set properly with either the business or operations and support users. This may be the first time they are exposed to a solution that “learns” over time. They also need to be trained to use a solution based on probabilities versus high accuracy business intelligence reports.
Legacy Technology
- Very few financial services companies are greenfield ventures. Established organizations bring with them legacy technology—not a great foundation for AI-powered solutions. Multiple inflexible systems, less than pristine content and outstanding technical debt all create friction in successfully leveraging AI.
Financial
- Fee compression and a prolonged low-interest rate environment combined with growing regulatory compliance and cybersecurity spend is hampering discretionary tech spend for incumbents. This puts innovation spending and large-scale projects at risk, particularly when the ROI is not clear across the entire organization.
Regulatory
- Wealth management firms tend to be more risk averse and have more compliance checks and balances than other B2C industries (e.g., travel and transportation, e-commerce, etc.), which are less regulated than wealth management firms. This can slow adoption of new technology.
Solving for X
How do companies dodge the hurdles and realize value from AI solutions? The answer starts with a question.
How do companies dodge the hurdles and realize value from AI solutions? The answer starts with a question.
What’s the What?
- What are the business problems the firm is trying to address and what are the goals?
- Is AI the right technology for the job? And what specific category of AI technology is the right fit solution for the problem?
Test the Foundation
- Are the CRM, core banking, investment books and records and other systems ready to support the AI initiative?
- Is there usable and interoperable content across the enterprise?
Look at the Culture
- Is there a transformation culture?
- Does it start at the top?
- Is there a steering committee to ensure that all stakeholders have a sense of ownership?
- Are the appropriate objectives and rewards in place?
Balance Short Term vs. Long Term
- Is there one business line that is well suited for near-term implementation—and success?
- Is the organization investing for long-term success?
- Who is the senior sponsor who will champion the project through budget rounds and business cycles?
In the coming years the typical human interaction with a computer will not be overwhelmingly dominated by the basic task of data entry. (Although anecdotal, see for yourself by classifying your technology interaction over the next 24 hours). As one looks to technology for useful decision-making insights and more contextual content, one’s fundamental expectations for machine interaction will evolve. Those enterprises that recognize this transformation and invest in developing organizational models to leverage the new paradigm will establish a significant competitive advantage versus those firms that do not adapt.
About the Authors
Roger Hu is a digital transformation leader who focuses on helping financial services institutions initiate and execute on client centric business and platform transformations. He is a Managing Partner at Gartner Consulting and founder of WealthTransformed. He has held leadership roles at Accenture and IBM Global Business Services and led transformative programs in industry at Morgan Stanley. Roger holds an MBA from NYU Stern and an AB from Amherst College.
LinkedIn: linkedin.com/in/roger-hu
Darren Duffy is a fintech advisor who specializes in the wealth and investment management sectors. He serves on the advisory boards of ForwardLane, Inc. and the Digital Innovation program at Brandeis University. He has held leadership roles at Refinitiv, Thomson Reuters, Lipper and Capital Access International. Darren holds an MBA from NYU’s Stern School of Business and completed the Securities Industry Institute at the University of Pennsylvania’s Wharton School.
LinkedIn: linkedin.com/in/darrentduffy
Twitter: @darrenxduffy
[1] For the purposes of this article, AI is defined broadly and encompasses machine learning and natural language processing.
