Stuart Grant is Head of Capital Markets, Asset and Wealth Management at SAP.
Discover top fintech news and events!
Subscribe to FinTech Weekly's newsletter
Read by executives at JP Morgan, Coinbase, Blackrock, Klarna and more
From fee compression to unfavorable shifts in macroeconomic conditions to mounting technology investments that have yet to pay off as expected, asset management organizations face significant headwinds as the calendar turns to 2026.
In a 2025 analysis of the global asset management industry, McKinsey & Company found, for example, that asset manager margins have declined by three percentage points in North America and five percentage points in Europe over the past five years as a result of factors like these.
But a pressure-relief valve is at hand in the form of targeted, well-placed deployments of artificial intelligence. AI in its various forms — generative, agentic, etc. — is beginning to demonstrate value in a range of front-, middle- and back-office use cases, giving asset managers the means to capture new productivity and efficiency gains, to identify and capitalize on profitable new business opportunities ahead of the competition. In its analysis, which is based on a survey of C-level executives from asset-management firms across North America and Europe, McKinsey determined that for an average asset manager, the potential impact from AI, gen AI, and agentic AI “could be transformative, equivalent to 25 to 40 percent of their cost base.”
The challenge for asset management organizations, then, is to determine where within their organizations AI can provide the most value.
Deploying AI for Maximum Impact
Companies across the asset management landscape are employing AI on a variety of fronts. Much of that activity is occurring within larger organizations that have the deep resources to develop their own capabilities around large-language models, with targeted AI agents and the like. But the other side of the AI coin is that it also can help asset managers outside the largest Tier One organizations compete on more equal footing against these larger firms.
What’s more, while many organizations focus their investments on customer-facing AI use cases, it’s important not to overlook the opportunities to create value with other scalable AI implementations across the front, middle, and back offices. Rather than seeking out point solutions that may not integrate well with one another, the wiser approach to generating value from AI could be to target investments that dissolve the virtual walls between the three office layers to create efficiencies, bolster productivity, streamline processes, and better inform planning and strategy.
In short, seek out AI uses cases that encourage — and can leverage — the freer movement of data throughout an organization. Here’s a handful that look especially promising:
1. Automate and speed financial close and other finance functions. Finance has historically been an area fraught with manual processes. With the help of AI agents, asset management organizations have an opportunity to automate many of the processes around the finance function, including the financial close as well as AR, AP, invoice reconciliation and the like. In these scenarios AI can support improved automation of data movement. It also can provide finance business users with proactive notification – and actionable scenarios – for potentially unseen issues with capital surplus/shortfalls, balance sheet adjustments and the like.
2. Improve risk management through true alignment with finance. Data from the back office can be immensely valuable to risk-management teams in the middle office. Those teams can use data around investor holdings, cash flows, market liquidity, margin/collateral, etc., combined with customer profile and communications data to identify early signals of client redemptions and associated liquidity risk.
3. Identify and quickly mobilize on opportunities for new fee structures and business models. Organizations can prompt their AI tools to research and model the impact of potential fee changes as well as new business models. What does historical data suggest about how a fee change would impact accounts receivables? Are there opportunities to split an existing area of the business (such as specific asset class or geographical funds) into two or more parts, or to bucket customers differently, and if so, how strong is the business case for moves like these?
4. Inform decisions about expansion into new products or geographies. Your organization is considering a move into a promising but relatively risky new geographic market. How have past moves like these turned out in terms of expected and actual costs? What are the likely regulatory and HR impacts of such a move? A dialogue with a generative AI digital assistant can yield valuable answers to questions like these, resulting in better-informed strategic decisions.
5. Model what-if scenarios around the potential impact of portfolio rebalancing on future earnings as well as customer investment priorities and risk appetites. AI tools can provide insight into the potential impact of these kinds of shifts, while also offering recommendations on optimal timing in light of accounts payable obligations and other factors. By making connections like this with data, AI helps to address information disconnects between the finance function and front-office portfolio management, supporting more on-point strategic-planning and budgeting.
In the case of one firm I work with, for example, they’re seeking to combine portfolio-attribution data on the performance of individual elements of its portfolio with data on customers’ risk appetite and fee structures. The goal is to better understand the financial reverberations of portfolio rebalancing relative to customer expectations and future earnings.
6. Heighten productivity. Some asset management execs I’ve spoken with recently say their organizations are looking to double assets under management without material increases in headcount, simply by leveraging AI and AI agents more broadly across their organizations. They’re creating AI agents and putting them right alongside employees — as digital extensions of those employees, essentially. Ultimately, the productivity gains these agents provide enable small and midsized firms to punch about their weight to compete on more even footing with larger firms.
7. Sharpen fraud detection during customer onboarding. AI is adept at rapidly scanning and validating the authenticity of onboarding documents, identifying even the most minor anomalies (in font size, document formatting, etc.) that can suggest a customer isn’t who they seem and thus require more screening.
As impactful as use cases like these can be within an asset management organization, maximizing their value depends heavily on the quality and accessibility of the data that feeds them. First and foremost, the data must be understandable to human and machine on a self-service basis. Oftentimes, firms pull data out of source applications and move it into a data lake. However, doing so removes vitally important semantics and context specific to the application environment. Without this metadata, AI’s output — and overall impact — could be suboptimal. So, organizations in many cases are better served leaving that data in its natural application environment alongside accompanying metadata. Think of the data in these applications as the batteries that power generative AI, agentic AI and intelligent analytics within an organization. The more powerful the batteries, the better positioned an asset management organization will be to leverage their AI investments to slice through the headwinds confronting them.
About the author
Stuart Grant is Head of Capital Markets, Asset and Wealth Management at SAP. For 20+ years he has been working with data in the capital markets industry in roles covering product management, business development and business management.
