AI in wealth management is moving past general-purpose chatbots towards agentic workflows that do the heavy lifting behind each client relationship. The quarterly client review shows what that means in practice: firms using agentic AI with connected data and content have cut review preparation from hours to minutes, which gives advisers more of their time back for the clients themselves.
I recently heard the head of a large US wealth management business talk about their client review process. The conversation turned, inevitably, to AI and how far it still had to go. I was astounded to learn that for each of their 10,000-plus clients, every review deck was assembled largely by hand by their adviser four times a year.
From an initial account stub, the adviser and their team tailor the report to the client’s prior meeting, prioritising certain portfolios based on importance and moving others to an appendix. Then they look over the client’s situation for anything that has changed. It might be a new job, a baby or retirement getting closer, and each one is an opportunity to improve the client’s future financial outlook and, importantly, to cross-sell.
High touch in private wealth is the name of the game, but when prep like this is manual, the context is the first thing to go. The adviser turns up with positions and performance but misses the opportunity to suggest a 529 plan (a US college savings account) for future college costs. The client sits through the review and leaves feeling that nobody remembered anything about them. For the firm, the downsides start with the operational risk of manual error and end in the lost revenue from missed cross-sell. When the tuition bills arrive in 17 years, the cost to the client is even greater.
How AI is used in wealth management client reviews
The talk was timely. Over the last year we have partnered with a number of firms in the wealth space to reimagine this client touchpoint through the lens of agentic AI and connected data and content. Firms using it have cut review preparation from hours to minutes.
The centrepiece is an agentic workflow that tailors content to each client’s situation. The adviser starts the process in Kurtosys, and the agent reviews the client’s life situation and adapts the content to suit. The adviser walks into the meeting prepared, with more time to win new business or look after existing clients.
Far from the free-form wild west of working directly in PowerPoint with a general-purpose chatbot, we provide an auditable, secure environment at every step of the process.
For the firms we have worked with, a consistent client data foundation is the bedrock. Accounting data sits in Snowflake, Databricks or Microsoft Fabric, and client goals, risk tolerances and life events come from the customer relationship management (CRM) system or customer data platform (CDP). Our platform reads all of it into a single entitled view in real time; messy extract, transform, load (ETL) jobs be gone.
The central reporting team then builds data-driven slides and product-related content into a single master in Kurtosys, linked back to the golden source of client data. Compliance reviews that content once, centrally, before it reaches any client. No slide is assembled by hand.
The adviser takes control from there. From their existing portal or CRM, the Kurtosys plugin lets them find the client, choose the report and work with the agent to tailor it. The agent finishes with its own compliance check and flags anything that needs a human to look at it. Compliance can adjust those rules and pull up any client’s past content at any time.
The review itself can run inside the tool, with anything new the adviser learns captured in real time and fed back into the CDP or straight into the report.
As more advisers delegate work to Claude and other agents, the Kurtosys Model Context Protocol (MCP) server lets them automate the week ahead. The adviser’s agent pulls appointments from Outlook and asks Kurtosys to prepare each review. The two agents agree an agenda between them, and the adviser checks the result.
What agentic AI means for wealth firms and their clients
Applied to those 10,000 quarterly client reviews, and the three to four hours of work that goes into each one, AI-powered tooling like this saves millions of dollars in adviser time alone. If just 1% of those improved reviews lead to better client outcomes, relationships retained or new products sold, the efficiency gains pale into insignificance.
Just as Uber and Airbnb brought once luxury items such as private chauffeurs and holiday homes into the orbit of individuals further down the net worth scale, so will AI have the same impact in the wealth space.
Far from commodifying the offering, AI will allow trusted fiduciaries to bring services once reserved for pension funds and family offices to clients who could never have justified the cost. Portfolios matched to each family’s own liabilities, structured products designed around one client’s portfolio and market view, and a family office’s grip on the household paperwork have always been far too costly to deliver below the very top of the market. With agents doing the heavy lifting, they come within reach of every client on the book.
The market opportunity feels significant and is a textbook case of Jevons paradox at work: make a service cheaper to deliver and people use far more of it. So the market for advice should grow as the cost comes down. There is a first-mover advantage to be had, though it takes portentous vision to invest in data and digital client channels now.
You will no doubt hear people in your business claim that your clients are too old to countenance this sort of technology. Take that view and you will be caught out sooner than you think. Meta is putting agents that shop on their users’ behalf into Facebook and WhatsApp this year, the very apps your clients use to keep up with their grandchildren. Once a retiree with a seven-figure portfolio has had an agent find and order a birthday present, they will expect the same from the firm that looks after their money.
The firms that win will be the ones already waiting when their clients get there.
If you’re thinking about how agentic AI could change your own client reviews, see how we support wealth management reporting and client portals, or book a demo with our team.
