Every wealth advisor knows that deeper client relationships tend to be both more profitable and more loyal (“stickier” in marketing terms). Therefore, deepening client relationships tends to be a priority for every wealth management firm, second only to new client acquisition. Client relationships are deepened by meeting more of your clients’ needs, so correctly identifying, even anticipating, your clients’ needs is the critical success factor.
Fortunately, wealth management firms have extensive data to learn from. Typically, this includes historical data on new clients, a relationship view of their products and services, profitability metrics, measures of relationship depth and growth over time, and transactional history. But making sense of all this data can be a daunting challenge. How do you know which data is important, or even relevant? Which data features are indicative of a need that you can fill? Are there trigger events that are predictive of an emerging need?
In the absence of a systematic approach to learning from your data and arming your wealth managers with insight, you must rely on each of your frontline teams to do their best with whatever information they have at hand. Results will be inconsistent at best. Machine learning offers a better solution, allowing you to use a number of different methods to identify, and even predict, your clients’ needs.
First, if you combine what you already know about your clients with data on the products and services each client has, then you can use machine learning to build a predictor of need. For example, if clients A, B, and C all have a need for product X, and client D looks very similar to clients A, B, and C, then it’s a good bet that client D needs product X, too. This is a classic supervised learning problem: You train an algorithm based on known outcomes (clients A, B, and C have product X), and then use the resulting model to predict the need for product X among the clients who do not have that product.
You can use this insight to:
Create client-specific prioritized need lists for your wealth advisors to use when meeting with clients. Combine the results of models for each product and service then sort by highest likelihood of need.
Create marketing plans or client outreach programs for each product or service. Use the results of the model to identify the population of clients most likely to have that need.
Deepen product expertise, training, and delivery capability for those products with the highest number of existing clients with a high probability of having that need.
Machine learning can also be used to look backwards to identify trigger events which presage a need. By looking at the population of clients that began using a new product or service and examining changes in their transactional patterns or demographics over the prior six to twelve months, you can use machine learning to build a model that predicts the emergence of new needs. Typical triggers may include life events such as births, marriages, retirement, inheritance, job or company changes, and asset purchases or sales. It’s a practical impossibility for a wealth manager to stay on top of life changes for every client, and to know which product or service these may indicate a need for. Machine learning can create models that alert wealth advisors to a change in their clients’ situation or behavior, which may indicate a need for a new product or service.
A seasoned wealth manager may be able to do all of this effectively using deep knowledge gained from long experience with all of his or her clients. Even then, he or she may miss subtle clues that indicate a need. With machine learning, models can be used systematically to drive client awareness, analyze needs, and identify deepening opportunities. These capabilities can then be leveraged consistently across your entire enterprise. This translates to improved advisor productivity, more satisfied clients, and higher profitability. A win, win, win.
About the Author:
H.P. Bunaes leads the banking practice at DataRobot, helping banks leverage AI and machine learning for predictive analytics and data mining. H.P. has 35 years experience in banking, with broad banking domain knowledge and deep expertise in data and analytics. Prior to joining DataRobot, H.P. held a variety of leadership positions at SunTrust, including leading the design and development of the risk data and analytics platform used enterprise wide for risk management. H.P. is a graduate of the Massachusetts Institute of Technology where he earned a Masters Degree in Management Information Systems, and of Trinity College where he earned a Bachelor of Science degree in Computer Science and Mechanical Engineering.