Wirehouses use predictive analytics to shore up goals-based planning

Corporations with big digital footprints, such as Amazon and Netflix, can predict what their customers will want to buy and watch by mining data and generating recommendations for individuals in real time.

Brokerage firms like Merrill Lynch Wealth Management, Morgan Stanley Wealth Management and Wells Fargo Advisors are putting their own spin on these sorts of predictive analytics to augment how their advisers work with clients and ultimately deliver financial advice.

“Every big corporation is looking at this right now,” said Kelley Byrnes, a wealth management analyst at Celent, a technology-focused research and consulting firm. “Definitely all the large brokerage firms.”

PREDICTING THE FUTURE

At its core, predictive analytics is a strategy using past and real-time data, trends and similarities among clients to predict likely future scenarios, Ms. Byrnes said.

One arena where the wirehouses are focusing their attention relative to predictive analytics is goals-based financial planning, in which advisers develop wealth management strategies focused on helping clients reach specific goals.

(More: How to take half the year off and still grow your assets by 20% annually)

While the fusion of technology and this type of planning is a few years old, broker-dealers are enhancing their tech platforms to get a better picture of clients and their communication preferences, to help advisers deliver better advice and client experience.

Wells Fargo Advisors, for example, currently uses client data, such as age, risk tolerance, net worth, liquidity needs and investment time horizon, to create a client profile and send relevant messaging to advisers about certain clients, said Zar Toolan, head of advice quality at Wells Fargo Advisors.

For example, advisers may receive a message from the firm reminding them to start a discussion around required minimum distributions with a 69-year-old client, Mr. Toolan said.

FORWARD-LOOKING

Wells Fargo is currently developing technology to make the process more forward-looking. In other words, rather than providing a snapshot of where clients are, the idea is to help figure out where they’re going, in the same kind of way one receives directions through Google Maps, he said.

“That, from the perspective of helping people to get where they want to go, is the most exciting part of predictive analytics,” said Michael Liersch, head of behavioral finance at Merrill Lynch.

A client may not have articulated certain goals or may not have considered a specific one, for instance. Predictive analytics can help advisers pinpoint something that may have gone unsaid by a client, and allow them to proactively introduce the idea and develop a wealth strategy, Mr. Liersch said.

MORE DYNAMIC

“It’s a much more dynamic understanding of who this person is,” said Denise Valentine, a senior wealth management analyst at Aite Group. “It’s a much more sophisticated tool that can theoretically predict [a client’s interests] better.”

Cheaper data storage and processing costs, a tough market environment and pressure for big brokerages to differentiate their services as industry consolidation creates more homogeneity are some of the factors driving such tech changes, Ms. Valentine said.

The Department of Labor’s fiduciary rule, which raises investment-advice standards in retirement accounts and is set to come into force in June, has also created an impetus for firms to invest in new predictive-analytics technology to better understand customers, according to a 2016 Celent report.

Historically, the business has relied on advisers’ astuteness and “gut,” but that is not so much the case anymore as predictive analytics helps them make better decisions, Ms. Valentine said.

“No matter how good you are as a human, you’re going to miss stuff,” said Jeff McMillan, chief analytics and data officer at Morgan Stanley.

(More: 10 surprising adviser tech trends)

Machine-learning technology is enabling firms to develop algorithms that learn, adjust and change in real time based on clients and marketplace dynamics, he said.

In other words, unlike today, firms don’t have to develop new algorithms to account for new variables.

MACHINE LEARNING

For example, such technology can tell if a piece of digital content for clients resonates, and can automatically increase the targeting and frequency of the content, Mr. McMillan said.

Morgan Stanley is developing the next generation of its first predictive tool, called Insights Engine, which will incorporate such machine-learning technology.

And it will have an updated communication component that tells advisers how clients prefer to interact with advisers: the medium (phone, email and text, for example), time of day and frequency of message, Mr. McMillan said.

An end goal, for which Mr. McMillan didn’t specify an implementation time line, would be to arm advisers with customized messaging and recommendations in line with a customer’s expectations in response to a major market-moving event like the next Brexit, he said.​

“Everyone is moving in that direction,” Mr. McMillan said.