Using AI to scale wealth management

Using AI to scale wealth management

February 10, 2022

Darya Shmat

Banking & Financial Technology Consultant

At this point, it’s safe to say that artificial intelligence holds the title of the most discussed technology across industries. At the same time, the technology’s popularity doesn’t correlate with the pace of its adoption. While wealth management firms have long been aware of possibilities AI offers, most of them are still uncertain if the game is worth the candle.

However, the need for change in the wealth management sector has reached its climax. Faced with intense competition, increasing customer demand for digitized experiences and fee reduction, and the avalanche of new investment opportunities, companies need to find new ways of engaging clients, generating leads, optimizing work, and standing out in the market. On top of that, with the majority of operations forcefully going digital due to the pandemic, many firms are struggling to find new clients and retain the existing ones.

The recent BFSI digitization has prepared the ground for wealth management companies to harness AI software development. With more and more organizations tapping into AI, now is the time to move from technology exploration and pilot projects to company-wide implementation. 

AWM firms struggle with mastering new-age competencies

To understand why now is the perfect time for wealth management companies to reinvent themselves, let’s recall how this sector has developed over the past decade. According to Statista, from 2009 to 2020, the wealth management sector has doubled the value of its assets, growing from $45.6 to $103.1 trillion. This can be explained by low-cost products finally gaining market share, growing middle-class affluence, and developing economies’ shift from addressing needs to satisfying wants. It’s worth noting that unlike many other business sectors, the wealth management industry has managed to grow by 11% despite the global pandemic.

AUM growth in trillion U.S. dollars

However, extracting value from these opportunities is a great challenge. According to  the 2020 Accenture report, 55% of wealth management companies expect less economic stability. In this context, wealth management firms need to be able to address sudden short-term market shifts while also keeping track of long-term opportunities.

Companies expect less economic stability

Most impactful AI use cases

Whether it’s market forecast, banking personalization, manual work automation, or machine learning-based fraud detection, with a carefully tuned model architecture and sufficient data quality, AI can solve the majority of wealth managers’ challenges. With 78% of organizations already deploying both client and advisory-facing AI-driven technology, it’s a serious test of their digital transformation capabilities and a catch-up game for the other 20%. Let’s discuss exactly how artificial intelligence can help wealth managers optimize workflow efficiency and drive more revenue.

AI gaining traction among AWM firms

Lead generation

Until the advent of augmented analytics and AI, wealth managers had to rely on manual data acquisition and analysis to find potential clients. In that case, decisions were mostly based on conventional metrics like client demographics and net worth. With AI, wealth managers can micro-segment their prospects based on a wider range of data sources including social media, niche news stories, and various public data sources, find new leads, and tailor pitches to them.

Furthermore, an AI system can help companies to connect prospects to relationship managers that share the same interests, are in the same age group, or have had similar clients in the past. For example, Finantix, a California-based financial technology provider, developed the AI-driven technology that can mine LinkedIn data to see if the relationship manager is already connected to the potential client and generate a pitch message in the appropriate tone.

While it may seem that formulating an attractive offer is something that only humans can do, Chris Burke, Vice President at RBC Wealth Management, explains that AI-based technologies like NLP can process large  amounts of both structured and unstructured customer data and quickly optimize conversations based on details from prospects' profiles. Clients’ risk tolerance can also be more accurately determined by assessing how their transactions change in response to market events. This way, the technology can significantly increase wealth management companies’ chances of winning new clients.

Fostering customer relationships

In the context of wealth management and financial advisory, establishing meaningful connections with your clients is the key to success. We have already entered the new era of clients demanding an increasingly wider range of services and hyper-personalized financial guidance underpinned by flawless user experience.

With AI-powered employee-facing robo-advisory systems, wealth managers can predict what next actions are best in terms of satisfying customer needs. By delivering more meaningful and personalized communication, wealth management firms have a much higher chance of increasing customer loyalty and retaining clients long-term.

For example, the Morgan Stanley Wealth Management Unit developed a Next Best Action system to help financial advisors match investment possibilities to client profiles.

Jeff McMillan, the company’s chief analytics officer reveals that the system’s advanced AI algorithm allows advisors to generate investment offerings much quicker and with greater precision. Importantly, McMillan said that the real value of such a system lies in its ability to identify clients’ topics of interest and enhance customer engagement.

The rapid influx of AI-based fintech also initiated a shift towards reduced fees of financial advisory, which calls for adjusting pricing models based on clients’ investment profiles rather than service quality. For example, the online trading platform Robinhood puts zero commission pricing models as its unique selling proposition. In a  wealth management context, the implementation of flat-fee models requires a granular understanding of clients’ profiles and accurate forecasts of returns on their investments.

On top of that, a carefully tuned predictive analytics system can help detect clients with a high attrition probability. This way, firms can determine these clients’ pain points and take preemptive measures to make sure they stay with the company.

Financial advisory automation

In 2020, robo-advisory platforms and other tools to analyze the stock market with machine learning surged in popularity, which can mostly be attributed to the pandemic minimizing physical interaction and causing financial volatility. For example, California-based automated investment service Wealthfront reported a 68% growth in account sign-ups amidst the pandemic.

Notably, Wealthfront’s robo-advisory platform is among the few that provides digital-only financial planning and investment management services. Wealthfront’s underlying AI algorithm analyzes a client’s saving and spending patterns and automatically determines the optimal steps for reaching their financial goals.

End-to-end decision-making automation has generated a lot of interest throughout the years, but couldn’t really earn clients’ trust. This is why in 2021, Wealthfront decided to adjust their robo-advisory platform and put more control in the hands of investors to maintain long-term customer relationships.

Vanguard, on the other hand, also deployed an automated robo-advisor platform, but no actions are taken without the confirmation of managers and clients. Importantly, Vanguard has become one of the biggest players in the robo-advisor realm with over $221 billion in assets under their management. This largely favors our argument that in the wealth management context, AI shouldn’t replace humans but assist them instead.

AI to augment humans in AWM

Back-office automation

According to  recent McKinsey research, relationship managers spend up to 70% of their time on advisory-irrelevant activities. This is due to wealth management companies still relying on manual data analysis for asset recommendations, risk and compliance analytics, as well as  lead generation. 

RMs spend 70% on non-advisory activities

Implementing AI, companies can automate many tedious and repetitive back-office operations, enabling managers to focus on more value-adding activities and become more productive. For example, the conventionally manual approach to KYC is notoriously error-prone, cumbersome, and inefficient. AI-powered data-extraction tools like Magic DeepSight allow for up to 70% reduction in costs associated with manual KYC data analysis. Similarly, AI tools can be applied to automate reconciliation, invoice processing, and fund accounting. 

AI implementation challenges and their solutions

Despite the huge potential of AI in wealth management, only a few companies have been able to apply this technology at scale and make it a functional part of their enterprise. Let’s discuss two main barriers on the path to adoption.

Top challenges for AI adoption in wealth management

Data governance

According to a recent PwC study of AI for asset and wealth managers, many firms are reluctant to scale AI because they aren’t sure about the technology’s reliability. This is understandable, since data privacy remains a top concern in the wealth management sector, further amplified by increasingly rigid regulatory requirements. Indeed, a poorly tuned AI model will most likely create more risks than opportunities.

Since the AI model output can be as good as the data fed into it, the success of AI initiatives strongly correlates with the level of maturity of the corporate data management infrastructure. Wealth management companies should ensure their data is accurate and accessible and that the processes of data sourcing and analyzing are aligned with regulatory requirements. Unfortunately, given that many wealth management firms have relied on manual data collection for decades, gaps in client profile information and lavish amounts of unstructured data can create significant limitations. This can be further exacerbated by siloed data repositories and the absence of a unified data platform.

To overcome these challenges, companies need to take a step back and revamp their data governance frameworks. First and foremost, this implies developing data standards and glossaries, implementing quality assessment tools, and establishing data governance roles. Further down the line, it’s paramount to establish data governance policies and controls, reporting frameworks, and automated solutions for data reconciliation.

Change management and new talent recruitment

Right after AI system reliability and data privacy concerns, most companies consider new talent recruiting, current employee retraining, and change management as their next most challenging AI adoption tasks.

Regardless of your company’s AI adoption stage, it’s important to let the workforce know about the incoming changes as early as possible. For example, assembling multi-disciplinary teams for AI projects makes a company’s strategic intent clear for the rest of the organization. It’s also important to start with AI use cases that are most feasible to demonstrate the technology’s real-life value.

For example, the aforementioned company Morgan Stanley, first started with a rule-based system to suggest investment offerings. Initially, this system had nothing to do with AI, but it showed where the company was  heading and helped them to get quick returns from automation.

In general, back and middle-office automation use cases are  a great starting point for the majority of wealth management companies. With early adopter firms already paving the way for AI-driven back-office automation, other companies can use them as an example and learn from their mistakes.

Workforce reskilling and talent recruiting should also never be an afterthought. It’s important to develop training programs and identify missing roles as early as possible. Given that there is currently a shortage of AI talent, it’s critical to develop a long-term talent strategy to fully benefit from AI in wealth management. Hiring new talent can be particularly difficult since candidates ideally need to have domain-specific knowledge in both technology and finance. This is why companies should bridge the gap between IT and business development teams and first look for talent internally. Besides, here HR predictive analytics may also come in handy, as such AI-powered solutions enable organizations to quickly evaluate their existing workforce and find and filter candidates with relevant domain expertise.

Conclusion

Undeniably, artificial intelligence is every wealth manager’s goldmine. While the competition from early AI adopters is rather tough, wealth management firms should not approach AI implementation head-on. AI calls for thorough preparation and, most importantly, for every part of the organization to be on the same page. To capitalize on opportunities offered by this technology, wealth management companies need to objectively assess their market position, long-term goals and technological readiness and develop a detailed implementation roadmap.

In the end, those who manage to overcome initial barriers and make AI a vital part of their companies’ workflows have all the chances to end up as market leaders.