Data Management, Analytics, AI and ML: Their Roles in Elevating the Wealth Management Proposition in Asia
Aug 30, 2023
Data is the new oil for the wealth industry, an expert at a Hubbis live event pronounced way back in 2018. But during those heady and uber-optimistic pre-pandemic years, it was more of a concept, and in reality, there was little realisation of just how much money and time would soon be spent on digitisation in general and then on AI and ML technologies. Fast forward to 2023 and there is already a dramatically different landscape – data is indeed proving to be the new oil, but more importantly, the wealth industry now understands in far more depth that it is the refining and upgrading of the data via ML and AI that can truly add value, relevance, personalisation, and ultimately effectiveness and better business. On July 20th, 2023, Hubbis hosted a virtual Digital Dialogue discussion in which the panellists confirmed these messages and at the same time explained how the application of ML, AI and other digital protocols and solutions are starting to revolutionise the offerings of asset managers, private bankers, EAMs, MFOs and even IFAs across the broad wealth management market in the region.
- Wealth Dynamix
- David Wilson, Principal Director, Asia Wealth Management Lead, Accenture
- Jitendra Tekchandani, Executive Director, Customer Science & Segment, Wealth Management, DBS Bank
- Andrew Bresler, Deputy General Manager, InvestCloud (APAC), InvestCloud
- Ronald Yim, General Manager, Hong Kong, StashAway
- Darell Miller, Managing Director APAC, Wealth Dynamix
Setting the Scene for the Discussion
FinTech Broadridge Financial Solutions produced a survey that highlighted how Data Management and AI Automation Tools are the top investment priorities for financial firms in the US, with 57% of respondents reportedly confirming they have some long way to progress before they reach what might be considered advanced stages of innovation.
The survey was compiled from a survey of 200 financial services professionals, conducted at a SIFMA conference in 2022. SIFMA is the leading trade association for broker-dealers, investment banks and asset managers operating in the US and global capital markets, and as such is not directly or solely related to wealth management, but the findings for this broad array of constituents can quite fairly be used to inform us on key trends and priorities in the wealth management community.
“In today’s rapidly evolving world, an optimized workflow is crucial and good clean data is key,” said Vijay Mayadas, President of Capital Markets at Broadridge, was quoted as saying. “And yet, firms are drowning in the complexity of managing and simplifying data, without the technology and digital infrastructures in place to support its management, stifling transparency, agility and growth.”
And it was the panel’s mission on July 20 to debate these exact themes around data, and how AI and ML can help refine and upgrade that data to in turn held drive a more client-centric and successful wealth management industry in Asia.
Expert Opinion - David Wilson, Principal Director, Asia Wealth Management Lead at Accenture: “Digitisation in the broadest sense is still a massive priority for the wealth industry in this region. Within that, data is absolutely crucial; it is vital to solve key issues around data gathering, data cleanup and then refining, and evolving strong data analytics capabilities.”
The experts first set the scene to provide the right context for the discussion, with the wealth management and Fintech/technology experts discussing the nature of the Asian wealth management market, defining the evolving business trends, and then identifying the data-centric digital solutions needs for the years ahead.
Everyone knows that the pandemic turbo-charged the drive towards digitisation, some of which was more catch-up and some of which was more forward-looking, proactive and market-leading. Accordingly, these experts, all leaders in their fields, drilled down to analyse where technology investment is focused today, from back-end to front-end, where it has been most effective, and where the newest and forthcoming AI and ML solutions will be able to significantly enhance the offerings.
Zooming in on the theme that data is the new oil and that it must be judiciously and efficiently refined and upgraded, the panel focused in particular on data, data management, data analytics, AI and ML. They debated why data is so important to personalisation and the wealth management journey, and what digital tools, protocols and approaches private banks and wealth management firms could adopt to enhance this vital area.
They also analysed whether technology investment is being properly targeted and executed, and considered how the different types of banks and other wealth management organisations competing in the Asia region can best select, adopt, and then assimilate these new data management and data leveraging technologies to the best effect.
The reality is that much time and money can be wasted by not taking the right approach to digitalisation and by making the wrong decisions. But on the other hand, they delved into how to articulate smart, targeted technology investments taken with clearly defined data-centric operational and business goals in sight – all with the private client’s needs and experience front of mind – and discussed how those will significantly boost the competitive positions.
The Hubbis Post-Event Survey – Views from the Market
Hubbis also asked delegates – all specialists in the Asian wealth management scene - for more detailed replies to several key questions. We have selected one standout reply for each question, as below:
Q: How can data management, analytics, AI and Machine Learning help to drive personalisation and enhance the capabilities of the RMs/advisors at private banks and independent wealth firms?
A standout response from a delegate: “First data management. Effective data management is the foundation for personalization. Private banks and independent wealth firms can use data management tools to collect and store customer data in a structured way. This allows RMs and advisors to access customer data easily and efficiently, enabling them to personalize their interactions with customers.
Regarding analytics, this can greatly help RMs, and advisors understand customer behaviour and preferences. By analysing customer data, RMs and advisors can identify patterns and trends that can be used to personalize their interactions with customers. For example, they can use analytics to determine which products or services are likely to be of interest to a particular customer.
And on Machine Learning, we believe this can help RMs and advisors personalise their interactions with customers even further. By using machine learning algorithms, they can analyse customer data in real time and make personalised recommendations based on each customer's unique needs and preferences. For example, they can use machine learning to suggest investment opportunities that are tailored to a customer's risk tolerance and investment goals.
Personalisation is central to all these missions. By combining data management, analytics, and machine learning, private banks and independent wealth firms can offer a highly personalised service and experience to their customers. RMs and advisors can use these technologies to provide personalized recommendations, customized investment portfolios, and tailored financial plans that meet each customer's unique needs.
Q: Can AI-enhanced data and analysis improve the wealth management investment and advisory proposition? If so, how and which types of competitors will most benefit?
The standout response: “AI-enhanced data and analysis can help wealth managers to offer better investment and advisory services to their clients by using data-driven insights, personalized recommendations, and automated processes.
It helps stimulate better decision-making: AI algorithms can analyse vast amounts of data and identify trends and patterns that humans might miss. This can help wealth managers make more informed investment decisions, and also help them identify potential risks and opportunities.
It helps with improved risk management - AI can help wealth managers identify potential risks and manage those risks more effectively. By analysing data in real-time, AI algorithms can flag potential problems and alert wealth managers before they become major issues.
It also enhances personalisation: AI can help wealth managers provide a more personalized service to clients. By analysing data about each client and their investment goals, AI algorithms can develop customized investment strategies that are tailored to the individual needs and preferences of each client.
And it also helps boost efficiency. AI can automate many of the routine tasks involved in wealth management, such as data entry, portfolio analysis, and performance reporting. This can free up wealth managers to focus on higher-level tasks, such as client relationship management and investment strategy development.”
Q: What are the major risks and limitations ahead related to the use of AI in wealth management?
A delegate’s detailed reply: “AI in wealth management can bring many benefits, such as better insights, personalization, automation, and engagement. However, it also comes with various risks and limitations, such as ethical, privacy/security, performance, control, and economic issues.
Regarding data privacy and security, the use of AI in wealth management involves collecting and analysing large amounts of sensitive client data. This understandably raises concerns about data privacy and security, and firms need to take steps to protect this data from cyber threats and unauthorized access.
There is a lack of transparency because AI algorithms can be complex and difficult to understand, which can make it challenging for clients to understand how investment decisions are being promoted or made. This lack of transparency can erode trust and confidence in the investment process.
There might also be bias and discrimination because AI algorithms are only as unbiased as the data they are trained on and the data they then utilise. If the data used to train the algorithm is biased, this can lead to discriminatory outcomes. Similarly, AI uses data that is available, and it is impossible to work out how much of that is unbiased, as there is an acknowledged lack of transparency around sourcing. Wealth management firms need to be aware of this risk and take steps to mitigate it.
It is dangerous if there is an overreliance on AI. While AI can be a powerful tool, it should not replace human judgment entirely. There is a risk that firms may become over-reliant on AI and fail to take into account other important factors, such as market conditions and client preferences.
As to the regulatory and compliance issues, the use of AI in wealth management is subject to regulatory oversight, and firms need to ensure that they are in compliance with relevant regulations. This can be challenging given the rapidly evolving nature of AI and the lack of clear guidance from regulators.”
The Key Insights & Observations from the Experts
In the upper client segments of wealth management, the RMs must be empowered with technology for better outcomes and to achieve scale
A speaker opened his observations by observing how vital mined and refined data are for the wealth market’s widely acknowledged mission to enhance the capacity and capabilities of the RMs. If RMs are empowered with technology and the right processes, the theory is that they not only have the resources and time to win more clients but also a greater share of the wallet from their existing clients.
“Helped greatly by machine learning, and then AI, you can drive personalisation for the clients and at the same time enable all your RMs to provide the same type of high-quality service and focus that most RMs today only deliver to their top 20% or maybe 30% of clients who produce the business,” he explained. “The banks and other wealth managers can thereby really scale their business, scale their profitability, but without scaling costs.”
The second key theme is content-led wealth management. Another guest commented on the vital need to differentiate client experience through content rather than just products or advice, which are oftentimes seen as largely commoditised. “I think there is a major challenge with content distribution within the industry; imagine, for example, a private bank still producing 15 to 20 trade recommendations a day, or more than 3500 of those a year. How do they hit the right client at the right time with the right ideas? Well, again, data and refined data enhanced by AI and ML hold the keys to unlocking that potential.”
He added that, in his view, AI and ML will not weaken the role of the RMs and advisors but would elevate and strengthen them. “It is all about enhancing insights, relevance and personalisation and augmenting the roles of the client-facing bankers,” he explained. “Wealth management will continue to rely heavily on human interaction, trust, empathy, and emotional intelligence and AI and ML will be enhancing the individual approach. It is all about doing more with the technology available Yes, there will be some concerns about the impact, but the businesses and people who embrace it and take it forward will get the most out of it.”
Expert Opinion - David Wilson, Principal Director, Asia Wealth Management Lead at Accenture: “The starting point for the articulation of the value of refined data in wealth management is the RM/advisor. The RMs represent perhaps two-thirds or more of the cost base of the banks and other firms, and they are likely to be the primary driver for banks and wealth management institutions to achieve ambitious targets of doubling revenues over the next few years. The leaders in this industry in this region are expecting the RMs to deliver the lion’s share of that growth. We need to help them solve for this given the current levels of RM productivity are a challenge, and we think that Generative AI actually gives us a really important tool to help them.”
Technology as the catalyst for intelligent banking – combining enhanced client-centricity, and greater personalisation, improving outcomes, reducing costs and scaling up wealth management operating models
“Data accumulation and management is not new, but technology has fast-tracked the use of data, and the raw material has suddenly become much more useful for everyone,” another expert opined. “Data is knowledge and helps our businesses become more intelligent, which is why we describe it as intelligent banking or cognitive banking.”
He explained that intelligent banking traverses all of their business areas. “We focus on cognitive banking as one of the pillars to accelerate our business outcomes. We have a data-driven operating model where we identify the outcomes we want, whether those are financial or non-financial, the latter being customer experience, satisfaction and so forth. Importantly, we believe focusing on the financial outcomes alone is insufficient; we need a more complete approach.”
This more holistic approach results in improved and better-improved decision-making. A more complete picture helps the RMs, the bank and the clients. Additionally, it is important to try to harness AI-enabled hindsight and foresight. In hindsight, you can review and use the data to uncover the stages where performance might be less than the bank or firm would want, and regarding foresight, the data helps predict where they can better serve and engage with customers effectively.
He said that it is vital to empower and enhance the RMs, agreeing with a fellow panellist’s comments that the client-facing banks spend too much time in general on non-productive tasks that nevertheless need to be done.
“The data and AI can help us to increase the productivity of our RMs, and enhance their expertise,” he reported. “Across the retail customers through our private banking clients, the RMs are simply unable to engage with the numbers of clients they have, so historically, they would work very largely with those who produced revenues, using their judgement to select those out. But we have taken a far more data-centric approach to that process.”
Accordingly, he said they now have some 15,000 data points for each and every customer, using that information to pinpoint priority amongst clients for the RMs. “That way, we solve the ‘who to call’ problem, and the data helps us identify when best to contact each client for maximising the outcome of the engagement.”
He elaborated on these points, noting that RMs’ technical skills can also be improved via data and AI by giving them financial insights, customer insights, and behavioural insights, but it is the softer skills that must be aligned as well. “The knowledge that they used to take years to achieve can now be assimilated more rapidly, but behavioural skills are very important too, and it is vital to know how to approach clients,” he explained. “To help with this, segmentation is really useful, so they understand the wealth continuum from retail to mass affluent to HNW status. You need to be able to understand the expectations of each segment and the diversity of approaches required.”
Data also helps focus RM minds on the particular segments within the wealth management market
An expert remarked that customer demographics in Asia are changing rapidly, with wealth accumulated ever younger. “It used to be that millionaires were minted in their 40s and nowadays it is often in their 20s,” he reported. “Moreover, the types of assets they tend to hold are different, with far greater range and diversity, so you then need to delve further to arrive at micro-segmentation. In all these areas, data and AI are playing a major role in delivering hyper-personalisation at scale and without incremental cost.”
Data plays directly into the wealth continuum in Asia’s dynamic economies – the more you know, the more you can anticipate the evolution of demand and client expectations, and the more proactive you can be
A guest identified several key trends within wealth management that align with the key objectives around the use of refined data. One is the wealth continuum, or the need to service clients as they mature and evolve their wealth and their needs, perhaps graduating from retail or mass affluent clients to the HNWI level.
“Data is also central to a smooth transition from retail to mass affluent, then to HNW and UHNW status,” he said. “Any clients transitioning within these segments are supported better with the organisation’s data history and experience, thereby helping oil the wheels to achieve the seamless shift required.”
Data and enriched data can be a key to unlocking more client-friendly and less costly and frustrating onboarding, KYC and AML processing
A guest shifted his focus to the use of data for onboarding and compliance, noting that if every single team has access to the same high-quality data, which is easily seamlessly accessible internally, the onboarding times can be shortened, potentially from many months to a matter of days.
“We know of several banks that have transformed their onboarding processes from around 45 days on average to a single day!”, he reported. “And incredibly, we learnt of a customer bank that had onboarded an offshore client and his wife in less than three hours, start to finish, completing all checks and documentation.
Robo-advisors are experts at leveraging data to deliver a cost-effective retail and mass affluent targeted digital investment proposition
Expert Opinion - Ronald Yim, General Manager, Hong Kong, StashAway: “On the use of data in wealth management, at StashAway we take a data-driven approach to investing, leveraging macroeconomic data to minimise risk and maximise returns for every portfolio throughout economic cycles. We augment this investment approach with human oversight to ensure the algorithms are continually created, tested and monitored.
Regarding the opportunities for using AI in wealth management, we are optimistic about how AI will revolutionise the wealth management landscape. Our teams have already started exploring how AI could facilitate data analysis, improve risk management, or even draft code. The goal: to keep delivering innovative, user-friendly investment solutions to more people.”
But even if you understand and agree with all these insights and concepts, remember that data must first be properly mined before it can be smartly refined and then employed to help evolve the wealth management proposition
An expert cautioned that ML and AI only work ideally on the solid data foundations that firms first need to put in place. “That is crucial,” he stated. “Without the right data strategy and structure in the first place, ML and AI will not be effective, and you will not achieve the levels of personalisation and ultimately better service and scale that you seek.”
He elaborated on this, pointing to research on a variety of advantages that will flow from this approach. First, there is the automation of repetitive manual tasks such as risk assessment, portfolio rebalancing and data analysis, freeing up time for wealth managers to focus on complex personalised aspects.
Second, you can enhance decision-making using analytics and AI-driven algorithms to produce improved data-driven decisions based on the client's personal financial goals and risk tolerance.
Third, AI can help improve efficiencies and cost-effectiveness. Four, it will help achieve more tailored and personalised experiences at scale for clients, allowing wealth managers to better understand their preferences and needs.
And five, he concluded, AI support better risk management and regulatory compliance by analysing the vast amounts of data and identifying underlying potential risks and irregularities.
Expert Opinion - David Wilson, Principal Director, Asia Wealth Management Lead at Accenture: “The RM community requires augmentation. Far too much of their time – perhaps 70% or even more – is spent on non-revenue generating tasks. For example, servicing activities such as reacting to lots of client requests for reports, or helping with the basics of transacting, unless highly complex, are not really adding value, but the RMs spend perhaps 20% of their days handling just these such tasks. Additionally, RM productivity is central to the future, and there is still a lot more to be achieved in terms of enhancing their skills, capabilities and ability to scale. If we can strip away a huge amount of the administrative burden from RMs, then we are further towards being able to deliver true advisory-based sales to the clients.
To really make the RMs effective, digital tools and data, if applied well, can help enhance RM connectivity to clients, and help them understand their personal situations, expectations and preferences, as well as their actual behaviour. In doing so, the RMs will build more trust and be able to deliver a more complete set of services and solutions, potentially involving different arms of their banks, for example, corporate and investment banking as well to Asia’s many business-owner families.”
No matter how much theory you espouse, you must have the right data management and implementation strategies in place
A speaker then refined some of these observations, noting that you can have all the augmented and refined data in the world, but if you do not map and adjust your workflows to the new reality, the end clients will not ultimately see any benefits. “You must have the right approach so that you can really add value with these solutions,” he advised.
Another guest observed that the inefficiencies that exist within the wealth industry today are where AI can begin its improvements. He cited a McKinsey report that said data-driven advisors are set to deliver 15 times the value to clients than those that are not data-driven. It was also cited that 91% of all data is incomplete, outdated or inaccurate. “That is the starting point,” he said.
He pointed to the key HNW segment of wealth in Asia that they target for clients with USD2 million to USD10 million of AUM, in other words, just below pure private banking thresholds today, and above the wealth of the rapidly expanding mass affluent market.
“In this key segment, there is both a labour and skillset shortage in the region, with a dearth of experienced relationship managers,” he explained. “They also tend to serve somewhere between 100-200 clients, but only concentrate on 15% to 20% of that book. In this area, leveraging AI and tools that provide recommendations on which activity to prioritize is critical to the success and to gathering more share of wallet.”
Expert Opinion - David Wilson, Principal Director, Asia Wealth Management Lead at Accenture: “There is a finite number for the capex the banks and other competitors will be able to allocate to digital transformation, so they need to have their strategies in place and prioritise accordingly. The technology solutions capex has to cover a wide array of transformation topics, from onboarding/KYC to operational efficiencies, compliance, online protocols, apps, AI, GenAI and so forth. And each of these allocations needs to have a viable payback and timeframe to get on the lists. We are at the early stage of seeing significant funding being allocated to GenAI, but I expect to see strong growth in 2024 and beyond on two tracks – building out the operating model and infrastructure to support GenAI at scale combined with initial no-regret pilots to unlock value and generate learnings and momentum.”
He observed that the UHNW segment is more ‘white glove’ and with far fewer clients allocated to each of the very experienced RMs, but even there, they need as much support from data, driving insights and recommendations.
And in the rapid growth mass affluent market across the region, the mission amongst banks and other providers is to improve coverage, achieve scale, reduce cost-to-serve, and of course, garner more clients and share of wallet. “To achieve these goals there has been a widespread drive to deliver a pure digital experience and, at the same time, a desire to make that experience hyper-personal,” he reported. “And data is absolutely central to all that.”
Expert Opinion - David Wilson, Principal Director, Asia Wealth Management Lead at Accenture: “RM productivity is also vitally important because so many banks and others are hiring, and there is a limited talent pool. Even in Singapore, where there is a lot of experienced talent, the growth of the underlying market is such that the client numbers are expanding faster than the talent pool. Accordingly, the industry has increasingly been looking at hiring from outside the traditional sources and pipelines. This is likely to work better in the years ahead as wealth management, in our view, will be increasingly about more ‘softer’ and behavioural skills than simply about technical capabilities.”
Have faith - if carefully managed and properly implemented, AI-guided and ML-enhanced data could unlock a brighter future for the wealth management industry
A panellist rounded off his observations by reiterating his belief that in the wealth management industry, AI and ML will be empowering forces, and not destructive to the essentially personal approach of the past. He explained that it is about enhancing the individuals concerned, not removing them from the equation.
Another speaker reiterated the vital importance of a properly articulated strategy around data. “Too often, banks or other wealth firms do not have this,” he cautioned. “A valid data strategy consolidates and uses data in a way that builds that personalisation and that is fundamental to unlocking the power of AI and machine learning in wealth management.”
But there is a reality check ahead - the all-pervasive world of regulation and compliance teams’ wariness over the inscrutability of AI will very probably hamper adoption
He closed his remarks by nevertheless noting that the speed and momentum of change, which have been quite incredible in recent years, might slow somewhat in the foreseeable future, at least in terms of the implementation of AI and ML in wealth management.
“I do feel things will evolve more slowly in the foreseeable future, due largely to regulation,” he reported. “When I sit with decision makers and recount the advances in AI, ML and in other areas, their excitement rises, but so too do their concerns about governance and compliance around areas that are extremely difficult to comprehend. The worries naturally focus then on the ‘what if we get it wrong’ scenarios. In summary, the use of ML and AI for empowering RMs, yes, it will likely all advance apace.”
But he cautioned that there are major impediments ahead in the form of the human need to keep AI under control. He implied that direct-to-client AI-enhanced solutions will progress more guardedly as compliance gatekeepers will be increasingly wary of what they find hard to define and verify. In short, the devil you know might be preferable to another form of fiend masquerading as a saviour.