Now Reading:

An Overview of AI for Wealth Management – What’s Possible Today?

Private Equity	5758 VIEWS
Private Equity
5758 VIEWS

Thursday, January 31, 2019 10.17AM / By Raghav Bharadwaj of Emerj / Image Credit: Medium

According to a 2017 report by the Financial Stability Board, artificial intelligence (AI) and machine learning firms managed assets of over $10 billion in 2017, with further growth projected in the next five years. The reasons for this are clear; AI can now present wealth managers with new capabilities to enhance and further personalize their services, at scale.

The advent of big data and AI has changed how wealth management professionals go about their work. What was traditionally a “cookie-cutter” approach in which clients whose trading preferences would be categorized as “aggressive, conservative, or balanced” might now be more nuanced with AI. AI has the potential today to enable wealth managers and financial advisors to understand client requirements better. AI sentiment analysis can help with this on social media data, and customer information can help coax out insights about the current and future financial state of each client.

Identifying a major life event in a customer’s life, such as a retirement, might help advisors offer customized investment services. AI can be used to make sense of a customer’s social media interactions, transaction and investment data, and their declared preferences, to help advisors offer more personalized service to clients while simultaneously finding the right product-customer opportunities.

While AI can be used to make sense of a customer’s social signals, it can also be applied to analyzing a customer’s communication history. AI software can track and monitor a customer’s preferred communication channel (mail, email, phone calls, text messages) and what time of the day the customer prefers to receive communications. AI software might also predict the customer’s preferred frequency of communications and prompt wealth advisors to initiate any communications accordingly.

Additionally, regulatory compliances involved with wealth management or financial advisory services (such as MiFiD II or GDPR) are evolving. Ensuring that the returns for clients are maximized while still maintaining regulatory compliance might be challenging for wealth advisors with thousands of clients. Analyzing thousands of policy constraints while simultaneously finding the most profitable trading strategies for clients involves massive amounts of time and human effort spent in sifting through data.

Wealth advisors at large firms might see a reduction in costs and time by using AI software to monitor and track regulatory risk. AI software can today provide wealth managers insights and recommendations by taking into account a client’s preferences, financial trading trends, and regulations associated with advisory services at a speed and scale that cannot be matched by human analysts.

Moreover, AI systems “learn” to provide more accurate and personalized insights over time by using feedback from clients and by observing behavioral patterns in client data. We spoke with Robert Golladay, Managing Director, Europe at CognitiveScale, which offers AI software that helps both wealth advisors personalize insights and identify new opportunities for clients. According to Golladay, AI is being applied to wealth management services in two areas today:

Personalization: Helping financial advisors identify the investment preferences of a client and provide personalized advice to a degree that was not possible before. This might involve taking into account factors such as declared, observed, and inferred information around client goals or attitude towards risk.
Engagement: Helping wealth advisors communicate the most relevant insights for a client at a preferred time and channel.

Listen to the full interview with Robert Golladay here

The Changing Landscape of Wealth Management

Traditionally, when approached by a client looking to invest, a wealth manager might have made phone calls to the larger financial investment institutions to purchase research and access information about quantitative trading data. They would provide the wealth manager with a one-size-fits-all type of investment advisory service. With the advent of AI, wealth advisors could start to make sense of financial market signals and customer communication preferences to contextualize and personalize investment advice.

The last couple of years have also seen the rise of FinTech companies as competitors to banks, with the promise of better customer service and customer interaction options in wealth management. In addition, banks and other wealth advisors might also be pressured by a changing paradigm in customer communication preferences.

It seems as though the millennial generation-investors have their “antennae tuned,” so to speak, toward robo-advisors and FinTech. The fact that FinTech companies allow users to easily access investment ideas or research financial trends in news and social media through a companion app seems to have resonated with younger investors.

Robo-advisors and FinTech companies also seem to be focused on targeting the “low hanging fruit” clients with lower levels of financial resources. Banks and institutional investment firms might lose out on potential future clients to these newer entrants in the market unless their wealth advisors are equipped to deliver more personalized experiences to customers.

Augmenting the capabilities of wealth managers with AI tools might allow wealth management divisions at banks to provide personalized financial advice to clients at scale.

Banks and financial institutions might stand to benefit from using AI to effectively leverage data and deliver insights to human wealth advisors at the right time. This data might include user preferences, financial product usage, news, media, transactions, and other relevant data. AI techniques applied to such data hold the promise to inform investment decisions and provide wealth management insights that cannot be gained from traditional software.

AI for Personalization in Wealth Management

It might not come as a surprise to wealth managers that understanding the preferences of a client, their goals, and their current financial state are essential for providing the most accurate investment advice. Additionally, this data might enable personalization, thereby allowing wealth advisory service providers to differentiate their offerings.

Golladay says that AI can help wealth advisors make use of market trend information and data about clients or their investment preferences. He adds that AI software can also identify patterns in the data and understand the context in the data.

For instance, AI software might be capable of scouring through all the news media from the Financial Times or Bloomberg and automatically prompt wealth advisors with information about which of their clients might be at risk due to a particular market event, such as new financial regulation.

Golladay says there are usually five features to “augmented intelligence” software that can help wealth advisors personalize their services to each client:

AI can represent knowledge and identify relationships in the data within customer, market, or product profiles.

For example, a wealth advisor might use AI software to automatically predict changes to an investment plan for a client who is expecting childbirth based on the client’s current financial state and their declared goals.

AI can help wealth advisors with decision-making by correlating the time series data in each profile to deliver insights.

If a client had expressed interest in a particular company in past trading experiences, AI software might prompt the wealth advisors with trading insights on holding or selling stock for that company under volatile market conditions.

AI software learns through feedback and automatically improves its recommendations by incorporating preferences declared by the client, or by inference, trends from the profile data.

AI software might help wealth advisors identify correlations between financial events, such as the release of an annual report and a client’s investment choices, immediately thereafter. The software might then prompt advisors when similar opportunities arise in the future.

Financial advisory software is highly regulated and is mandated to provide proper justification for any advice. The recommendations that AI software for wealth management makes will also need to be highly explainable.

If a client was advised to make a financial decision based on insights from an AI software and ended up making a bad trade, the justification for said decision would be critical for client retention. In addition, the financial regulatory authorities in most countries mandate the need for explainability in any financial advice.

Assets and data can be used to create a specific AI solution.

In essence, wealth advisors might use AI software to create a “Profile-of-One” for each client. Through the application of machine learning and natural language processing (NLP) techniques, AI solutions might be able to generate these profiles for each client. The profile might include the client’s declared investment leanings and preferred characteristics identified from customer behavior data, such as transactional information on where and when the client spent their money.

Golladay explains with an example:

If Disney’s stock prices closed at below a 20th percentile relative to trading history in the last three months, AI software might firstly, recognize this event and secondly, recognize that the client has a preference for Disney and automatically pull up historical records for similar scenarios that happened in the past.

The software might find that this has happened three times in the past and had you acted within 7 days of the triggering of that insight, and then held the stock for 180 days, your median result will be a 27% return. You currently hold shares of NBC Universal Parks and held Disney in the past.

The software might then prompt the client with information on how that the tech company’s stock prices closed at similar prices three times in the past and suggest that holding the stocks for a month might lead to some median returns (in the form of a percentage). Golladay states that although data is being collected by wealth management firms, contextualizing insights from the data to each user might be the biggest value-add from AI software.

AI for Improving Customer Engagement in Wealth Management

With wealth managers coming under pressure to improve their customer-facing processes due to the advent of FinTech, understanding who to market their products to, in what language or through which channel (email, calls, messages and so on) might be critical. The fact that wealth management institutions usually have hundreds or thousands of clients makes this a highly challenging task.

Golladay gives an example of a wealth advisory firm which also has an asset management division:

Let’s say the asset management division created a new ‘organic mutual fund’ with carbon-neutral companies that are environmentally conscious and ‘green’. AI software can recognize that the financial product here has the attributes ‘organic’ or ‘green.’

The software can then scour through the advisor’s client base to find clients with a certain level of investable income, whose lifestyles indicate the label ‘organic.’ Lastly the software might prompt the wealth advisor to offer the fund to the prospective client at a particular time of day through the customer’s preferred channel of communication.

What business leaders in the wealth management space might need to note here is that information about the product and client lists can be combined to create query-able databases. Wealth advisory professionals can then search and discover the right product-customer fit.

For instance, AI software might “learn” to recognize over time events that lead to interactions with the client, such as a portfolio change recommendation, and automatically send communications to the client after similar future events.

Golladay points out that there may be several opportunities for AI software focused on self-directed investors. He adds that this might be especially true of wealth management AI software that provides these individual investors with the data and information resources measurable with those institutional investors.

Concluding Thoughts on AI for Wealth Management

AI software for wealth management might help businesses and individual wealth advisors better leverage data, such as customer social media interactions and investment preferences. Wealth advisors can gain insights about clients’ financial leanings, their level of aversion to risk, their current financial situations, and their intended future financial goals.

Wealth advisors at institutional investment services firms might find that using AI software might also help improve their customer engagement levels. By looking at historical data about customer interactions, the software can prompt advisors at the right time and recommend the channel of communication.

As Golladay puts it:

In the next two decades, roughly 30 trillion in financial and nonfinancial assets will pass on from baby boomers to their children. The winners in the wealth management space are probably going to use AI to understand how a person wants to be communicated with and marketed to.

In terms of real-world business benefits, AI systems might help wealth managers and financial advisors in the following ways:

Scale: handling a much higher number of clients simultaneously by automating a number of steps in the advisory process. This essentially boils down to being able to increase assets under management.
Personalization: creating personalized marketing and advisory services for each client by analyzing their communication preferences leading to improved client retention rates.

This article was sponsored by CognitiveScale, and was written, edited, and published in alignment with transparent Emerj sponsored content guidelines.

About The Author

Raghav Bharadwaj serves as Content Lead at Emerj, covering our major industry areas and conducting research. Raghav has a personal interest in robotics, and previously worked for research firms like Frost & Sullivan and Infiniti Research.