The global asset management industry is at a critical juncture. Longstanding tailwinds —primarily in the form of low interest rates and stable GDP growth—have changed direction, compounding ongoing challenges presented by the shift from active to passive and traditional to alternatives. Together, these trends are forcing firms to discover more sustainable pathways to outperformance. After a decade of unprecedented market growth, industry costs have become increasingly sticky and revenues unpredictable. As a result, margins have declined by three percentage points in North America and five percentage points in Europe over the past five years. Against this backdrop, technology costs have grown disproportionately, yet this increased spending has not consistently translated into higher productivity.
At the same time, AI is emerging as a transformative force, with some asset managers starting to harness the technology to fuel the next wave of productivity. For an average asset manager, the potential impact from AI, gen AI, and now agentic AI could be transformative, equivalent to 25 to 40 percent of their cost base, according to our analysis. In our research, we identified pockets of AI-driven value in areas such as improving distribution flows, streamlining investment processes, automating compliance, and accelerating software development. And capturing these efficiencies represents only the first wave in what is likely to be a broader technology-led reimagination of the industry.
In this report, we explore a structured approach to achieving significant technology ROI. Drawing on practical insights and proven strategies, this approach focuses on identifying high-impact opportunities and establishing the foundational capabilities required to unlock sustainable value, including domain redesign, talent upskilling, governance and IT transformation, unified data platforms, and sustained change management. Our analysis is based on research on firms representing 70 percent of global assets under management (AUM), and on interviews with CEOs, COOs, chief information officers (CIOs), and chief technology officers (CTOs) from leading asset managers in the United States and Europe (see sidebar, “About our research”).
Seeking and not finding ROI on technology investments
Over the past decade, positive performance and net flows in asset management have been largely driven by market tailwinds, in particular, low interest rates, stable GDP growth, and geopolitical calm. However, since 2022, many of these supporting fundamentals have reversed. After a decade of unprecedented positive market performance and record levels of AUM, the industry experienced a sharp 10 percent decline in AUM in 2022, and while markets and flows rebounded in 2023, industry costs have been growing and becoming increasingly resilient, while revenues have been unpredictable.
The result has been margin compression, with pre-tax operating margins declining by three percentage points in North America and five percentage points in Europe between 2019 and 2023. North American asset managers, for instance, saw an 18 percent increase in costs over the five-year period—outpacing revenue growth of just 15 percent in the same period. Against a backdrop of inflation, volatile rates, and geopolitical instability, revenues have become increasingly unpredictable. In the face of these challenges, structurally managing cost has become critical to restoring asset management profitability and building resilience for future growth.
Solow’s 1987 observation, “You can see technology everywhere except in the productivity statistics,” rings especially true in asset management today. Technology spending has been a significant driver of rising costs in the asset management industry, far outpacing spending in other functions. Over the past five years, technology investment has surged by 8.9 percent CAGR in North America and Europe (Exhibit 1). This growth itself is warranted: technology, while always a central pillar of strategic transformation, has increased even more in relevance, as a lever for both productivity and growth. However, given the industry backdrop and productivity paradox, industry leaders are increasingly asking how they can capture value and a better ROI from tech investments, and what role AI and gen AI specifically will play in this effort.
Why asset managers struggle to unlock the full potential of technology
Despite increased technology investments, cost as a share of AUM—a key productivity metric—has remained relatively flat at the industry level. Moreover, operational expenses in other functions have not contracted despite the expectations that technology would create efficiencies. At the firm level, our analysis shows that asset managers investing more in technology are not consistently more productive than peers across key KPIs such as cost-to-AUM ratio (Exhibit 2) and revenue per full-time equivalent (FTE) (Exhibit 3). In short, while the data is noisy, there is no clear correlation between higher tech spend and improved productivity. In fact, while the trendline is slightly positive, the R2 value (or coefficient of determination—a statistical measure that indicates how well a statistical model predicts the outcome of a dependent variable) is 1.3 percent, suggesting there is virtually no meaningful relationship between spend and productivity.
What is the disconnect between technology spending and ROI? Productivity gains in asset management have remained elusive largely because firms spend more—sometimes significantly more—on maintaining operations and legacy systems, rather than on future-focused transformation. In our research, we found that due to the complexity of these systems, asset managers allocate on average 60 to 80 percent of their technology budget to run-the-business initiatives, leaving only 20 to 40 percent for change-the-business operations. Furthermore, of the change-the-business operations, just 10 to 30 percent (equivalent to only 5 to 10 percent of total tech spend) is directed toward firmwide digital transformation, while the remainder largely supports individual use cases that fail to scale and drive impact.
For one leading asset management firm with more than $1 trillion of AUM, roughly 80 percent of its technology spend went toward run-the-business projects. In 2020, faced with increasing margin pressures and significant levels of technology debt, the organization embarked on an end-to-end transformation to update its capabilities and reprioritize the bulk of its technology spend to change-the-business initiatives. As of the first half of 2025, the firm now dedicates 70 percent of its technology budget to change the business. They achieved this turnaround by strengthening core capabilities where they have a right to win (as opposed to getting distracted by non-accretive innovation efforts that previously consumed a disproportionate share of resources); transitioning to cloud-based platforms throughout the technology stack; adopting accelerating product development schedules of three- to four-month cycles versus the previous nine- to twelve-month cycles; and restructuring talent to reduce reliance on third-party contractors.
The tech ROI challenge is especially acute in asset management because most firms have fragmented systems supporting different asset classes. Asset managers also work within siloed data environments with no comprehensive, fit-for-purpose, front-to-back platform, making it difficult to integrate diverse data sources.
Many asset managers also rely on outdated and fragmented technology stacks, which drive up operational complexity and costs, while modernization efforts are often prolonged and expensive. And even after modernization, firms frequently fail to fully decommission legacy systems, resulting in bloated application portfolios and limited efficiency gains.
This dynamic creates a vicious cycle that has persisted for decades. As organizations continue to spend on maintaining legacy systems instead of modernizing, they build tech debt and pay a “complexity tax” in the form of time and money. This vicious cycle also exacerbates the gap between how CTOs and functional leaders in asset management perceive the value delivered by technology. This misalignment is often driven by technology that is not fit for purpose, as well as by siloed roles and divergent incentives that hinder shared accountability.
Many institutions are working to fully realize the impact of their technology investments, and some are already seeing meaningful results. For instance, in the banking sector, Singapore’s DBS Bank achieved 11 percent and 8 percent CAGR in net profit and revenue, respectively—outpacing the industry’s 6 percent and 8 percent—largely by taking an end-to-end technology approach grounded in domain transformation, intentional adoption, and operating model reinvention.
Leading firms recognize that AI is not just another wave of tech, but an opportunity to fundamentally rewire the institution and potentially transform the economics of business. This is enabled by next-generation AI systems that can learn, adapt, and act autonomously, embedding intelligence into day-to-day workflows and unlocking step-change productivity gains across functions.
The AI leapfrog opportunity
For asset managers, the AI revolution is a timely opportunity to break out of entrenched cost structures by increasing efficiency across business functions. More recently, with the advent of agentic AI, there is a once-in-a-generation opportunity for asset managers to recover and leapfrog profitability levels. Executed well, AI can help asset managers recover margin levels. For example, a mid-sized asset manager with $500 billion in AUM could capture 25 to 40 percent of total cost base in efficiencies through AI opportunities enabled by end-to-end workflow reimagination. To realize the value at stake, taking a role-based approach to automation by embedding virtual agents and traditional automation in seamless ways, alongside human roles, while focusing on change management and adoption, will be crucial.
On top of these productivity gains, some asset managers are seeing early benefits in both top-line growth and risk reduction through AI. Select use cases—such as optimized portfolio construction and more effective client targeting—are starting to generate revenue impact. At the same time, AI is helping to reduce operational risk through tools like automated compliance monitoring and the codification of institutional knowledge, which can mitigate material losses during talent transitions.
C-suite leaders at leading asset management firms we spoke with pointed out additional areas of AI-driven value creation, including improving distribution flows, enhancing data processing in investment management, automating compliance control, and transforming software development. While most firms are still early in the adoption curve, the potential for impact is becoming increasingly concrete across core functions. These early signs of value realization suggest that AI, when strategically deployed, can go beyond efficiency to deliver meaningful impact across the full asset management value chain (Exhibit 4).
In client-facing roles, gen AI is enabling more seamless and personalized interactions, and can have a 9 percent efficiency impact.1 Virtual assistants can deliver on-demand portfolio insights and help support relationship managers with real-time information tailored to individual client needs. Gen AI also supports automated onboarding, ensuring faster and more accurate data capture. On the content side, gen AI-driven tools are helping generate customized communication at scale, maintaining engagement while reducing manual effort.
In investment management, gen AI is transforming the way insights are generated and decisions are made, and can have an 8 percent efficiency impact, according to our calculations. Analysts are using gen AI-powered research assistants to synthesize data from earnings calls, financial reports, and conferences, accelerating the insight generation process. Portfolio managers are leveraging gen AI tools to refine strategies, narrow investment options, and optimize portfolio construction. Enhanced risk models and automated reporting are further supporting a more data-driven investment approach.
In risk and compliance, gen AI is streamlining previously manual and time-intensive processes, and can have an estimated 5 percent efficiency impact. Compliance officers now use gen AI assistants to interpret complex regulatory requirements and flag gaps in documentation. Gen AI-driven monitoring tools are being used to detect anomalies and flag potential noncompliance, enabling more proactive oversight. As operational workflows become more automated, the reliance on manual controls is expected to continue to decline.
Within technology, gen AI is reshaping how software is built and maintained, and can have a 20 percent efficiency impact. Developers are using gen AI code copilots to accelerate coding, debugging, and testing, significantly shortening development cycles. Gen AI-generated documentation is also improving consistency and knowledge transfer across teams. And in IT service management, gen AI tools are increasingly handling service requests autonomously, resolving issues quickly with minimal human intervention.
Taken together, these gen AI applications are not only boosting operational efficiency but also elevating insights and delivering a better experience for clients and employees alike.
Building foundations to scale value
Capturing 8 to 9 percent impact per use case as described above is significant, but only a start. To realize the full potential of AI and significantly improve the ROI on tech, asset managers will need to move beyond isolated efforts and take on domain-level reimagination and workflow rewiring with accompanying change management complexity. This is where the real scalable value lies and likely the single biggest failure point within asset managers. Past technology waves—such as cloud and advanced analytics—often failed to deliver expected benefits because firms treated technology as a siloed capability, pursued separately by asset class, function, or program, not as a strategic enabler embedded across the business. Unless these foundational gaps are addressed, impact will remain limited. Asset managers who act early and get it right will stay ahead of disruption and lead the industry with their ability to reinvest and innovate, leaving the rest struggling to catch up.
Through our research, we have developed an approach grounded in six core imperatives that will help fully capture the value of AI in asset management.
Domain-based transformation to unlock AI’s potential
Instead of pursuing fragmented use cases that produce incremental change, asset managers can reimagine organizational domains through zero-based, AI-enabled redesign of workflows. AI efforts should be anchored in strategic, domain-wide priorities—such as scaling new products or deepening regional presence—to unlock new opportunities as AI economics continue to improve. One top 30 asset manager that primarily serves US retail investors began its AI journey attempting to tackle hundreds of individual use cases—but failed to see the returns they expected. They then shifted to a domain-based strategy, focusing on end-to-end transformation of four high-potential functions: operations, marketing, distribution, and investment management. Each AI effort is overseen by a centralized office, with its own P&L and short-, medium-, and long-term ROI targets tracked by management. For example, the firm views marketing as an area in which cost benefits can be identified and captured quickly (for example, streamlining the request-for-proposal [RFP] process). Early efforts have delivered ROI and leadership expects more to follow swiftly.
Revamping talent strategies and operations for AI-driven transformation
As with any new technology, AI has implications for talent strategies, and firms will need to embrace organizational change to effectively integrate AI into operations. Engineering talent will need to be trained to build and maintain adaptive AI systems, while talent in non-engineering roles like relationship and portfolio managers will need training to use AI tools in decision-making. Depending on a firm’s starting point, the focus may be less on hiring new talent and more on upskilling existing talent and raising AI literacy—especially given the high cost and competitive demand for top AI talent. As companies develop AI-related skills, employees will become more versatile, capable of performing multiple roles, and less restricted by geographic boundaries, except where regulatory and compliance issues apply. In some departments, teams could be organized based on skills rather than traditional functions, enhancing flexibility and innovation. AI agents will become active collaborators, requiring new organizational functions—such as “HR for AI agents”—to define their hierarchies, roles, reporting lines, and collaboration models, much like HR does for human employees. This will expand the traditional scope of IT and accelerate enterprise transformation.
One top 10 asset manager had previously prioritized building employee capabilities in coding, but realized these efforts were no longer needed, given AI’s ability to generate and improve code. The firm shifted to building AI capabilities among its employees, rolling out an internal large language model chatbot that employees use for day-to-day tasks such as translation, document summarization, and generation of documents and email. The firm believes the value at stake is significant. For instance, it anticipates efficiency gains of roughly 70 percent with regard to establishing investment guidelines pursuant to an investment management agreement. Leadership estimates a savings of 100,000 hours annually for both query management and workflow automation.
Another leading asset manager expects to shift its talent priorities from coding to data engineering, to help prepare data and data architecture for integration with AI. Interestingly, the firm reports that it is their most senior and junior developers that get the most out of AI; senior developers use their extensive knowledge to get the most out of the new tools, while juniors unlock capabilities by filling skill gaps with AI. Lastly, in addition to upskilling their workforce, the asset manager believes senior leaders must also work to become more familiar with AI technologies and use cases, so they both gain the benefits of the technology in their own operating model, and fully understand the implications of AI for their organizations.
Optimizing operating models with AI to enhance efficiency
Among leading asset managers, a governance model blending centralized oversight with decentralized experimentation and delivery has emerged as the most effective approach. Firms are establishing central “control towers” to provide strategic governance, enabling tighter business-tech integration across prioritization, requirement definition, and outcome accountability. At the same time, individual business units are being empowered with the tools and autonomy to experiment and rapidly scale AI solutions. As automation flattens organizational structures and consolidates functions such as back- and middle-office operations, CIOs and chief digital officers will play a central role alongside business leaders in shaping the future operating model.
One top ten asset manager is reimagining its operations from scratch to become AI-forward, through a centralized task force made up of senior executives. To navigate this complex landscape, the organization has created a rigorous governance structure to oversee ongoing AI projects, which includes a committee of senior leadership that makes dynamic funding decisions on all rolling technology portfolio investments.
Maintaining control of technology road maps for competitive advantage
Leading asset managers will transform IT from an enabler into a competitive differentiator that unlocks productivity across the organization. As they undertake this effort, asset managers should retain ownership of their technology road maps, using vendors strategically while insourcing critical capabilities to enhance execution speed and ensure access to key technologies. A growing focus will be on adopting reusable AI “recipes” to standardize processes, reduce integration risks, and embed AI across the tech stack. This approach simplifies execution, lowers costs, and develops differentiated capabilities that are difficult to replicate.
A top ten global asset manager with a diversified offering spanning public and private markets and retail and institutional clients is focusing on reusable “recipes” and capability patterns to enhance efficiency and reduce risk in its AI strategy. After an initial period of experimentation in which the organization encourages its employees to test available AI tools, leadership identifies usage patterns and high-potential opportunities, which are then codified and embedded into processes. This approach has enabled the organization to focus investment on AI use cases that unlock the most value.
Another firm, a top 30 privately-held asset manager that primarily serves institutional clients, recognizes its relatively high degree of vendor dependence (a common situation in the industry). While vendors are bringing some AI tools to market, the asset manager believes these tools are not state of the art, and that the greatest value from AI will come from internal proprietary development. The asset manager aims to maintain control of its technology road map by protecting a core layer of proprietary data and layering on third-party solutions outside of this core layer.
Developing data strategies to realize value from AI
To address the challenge of integrating AI and decentralized data into the tech stack and ecosystem, asset managers will need to redesign their data governance practices. They must establish unified data platforms and implement robust governance strategies to manage unstructured data, ensure compliance, and navigate the risks around personally identifiable information in closed-source models. Leveraging knowledge graphs will be a key part of making data more contextual, accessible, and actionable, enabling more advanced use cases in automation, analytics, and personalization.2
The global head of asset management technology for a leading firm emphasized to us the importance of data strategy and governance in scaling AI capabilities. Rapid advancements have rendered cloud systems and data practices from a few years ago obsolete. While AI agents are expected to have significant impact, prioritizing data capabilities in change-the-bank budgets is essential to unlock the agents’ full value. Leveraging both structured and unstructured data—enriched with the necessary context for AI models—holds immense potential across all functions.
Enabling effective adoption of AI through cultural shifts and change management
Successful AI adoption requires gradual adaptation, structured support, and behavior rewiring. Learning effective AI interaction, such as prompt engineering, takes time, and initial results may be suboptimal, then improve with familiarity. Critically, the front line must take ownership of this “last mile” of value, engaging deeply in defining requirements and reworking processes to ensure adoption.
Specifically, firms must execute across a full set of change management levers to influence mindsets and behaviors:
- role modeling and leadership from senior personnel across the organization
- fostering understanding and conviction through clear messaging and communications
- offering training modules to upskill users and prepare them for change
- reinforcing with formal mechanisms (for example, incentives, awards)
A robust change management approach also requires a fully dedicated team (10 to 20 people depending on the size of the organization) responsible for implementing the aforementioned levers, in close collaboration with leaders across business units and functions.
Without these crucial initiatives, organizations will struggle to realize sufficient returns on their tech investments. Firms should invest in training and incentives that embed AI into daily practices and decision-making rituals. Many asset managers have taken the lead in building early AI capabilities and educating their teams. However, these efforts are often plagued by familiar challenges: numerous fragmented proofs of concept instead of zero-based redesign of processes, use cases launched without performance measurement in place, limited collaboration with business, and lack of focus on adoption. In our experience, for example, revenue efficiency gains from AI-powered software development life-cycle automation only emerge after teams move beyond initial tool usage spikes, with lasting behavior shifts and a 15 to 30 percent uplift typically taking six to nine months.
A top 30 asset manager expects to experience a certain degree of “pain” as it transforms, given the significant foundational work required before benefits begin to flow. This work includes financial investment, coaching for employees, and change management. To accelerate the process, the asset manager is focusing on adoption and accessibility through sandbox environments and a data marketplace that enables employees to experiment. The approach to adoption will vary by functional area, as some functions already have a technology road map and need less guidance, whereas others are starting from scratch.
For the asset management industry, embracing AI-driven transformation is no longer optional but essential. If effectively embedded into the organization, AI can address mounting margin pressures and unlock significant value. It offers asset managers a unique opportunity to rewrite the story around technology-related ROI and adopt processes and build capabilities that allow them to capture real value from their investments. However, doing so will require a step-change in how they approach these technologies.
Focusing on the six pillars of transformation detailed in this article is critical—we believe underinvesting in one pillar can topple the whole stack. With the control tower overseeing every step to ensure cohesion, asset managers can go beyond fragmented AI use cases to achieve measurable efficiencies and elevate client experiences. Those who act decisively and strategically will position themselves as leaders, while those who delay risk falling behind. Now is the time to reimagine how organizations work and harness the full potential of AI to future-proof operations and drive sustainable growth in asset management.
