Why Investment Groups Need a New Data Strategy to Unlock AI Agents
As artificial intelligence transforms the investment landscape, traditional data strategies are rapidly losing relevance. For investment groups, the future demands a fundamental shift - from focusing on how data is moved and transformed, to how it is understood and mastered. In this new paradigm, ‘entity’ management will be the cornerstone of competitive advantage.
What Are Entities And Why They Matter For AI
An entity is a uniquely identifiable object such as a portfolio, security, issuer, or benchmark. To unlock strategic value and enable AI-driven insights, these entities must be consistently defined, mastered, and governed across fragmented systems.
For investment teams in an AI era, the range of data entities to manage expands dramatically, encompassing not only traditional objects like portfolios, securities, and benchmarks but also model portfolios, clients, qualitative research, analytics and many more.
The Challenge Of Poorly Governed Data Entities
It is clear that the industry is in its infancy in many of these areas. Take model portfolios as an example. Despite their rapid growth and influence over trillions in capital, models remain scattered, inconsistently defined, and poorly governed - limiting their strategic utility. Similarly, the concept of a ‘client’ is treated as secondary in front-office investment systems, undermining potential for hyper-customisation and AI-driven insights at scale.
Why Investment Analytics Require A Mastering Structure
Other critical concepts, such as ‘analytics’ increasingly warrant a new level of mastering. For instance, a common metric like volatility varies widely based on its calculation methodology, assumptions, and data sources. To balance maximising AI’s potential while ensuring strong governance, a mastering structure over investment analytics used across different processes is critical.
The Decline Of Legacy Data Structures
Historically, data strategies revolved around architecture diagrams, ETL pipelines, and orchestration tools - essential in an era when connectivity was complex and bespoke.
But today, AI-assisted engineering, cloud-native platforms, and low-code tools have dramatically reduced those costs. What remains scarce and increasingly valuable is the ability to master entities across different systems and data sources.
Yet much of the industry still clings to legacy constructs like IBOR, ABOR, and the security master. Many strategies are also predicated on legacy 'closed-architecture’ systems, with restrictive licensing and data flows. Such constructs and systems quickly become liabilities in an AI world. Licensing models that restrict usage, sharing, or AI training will inevitably stifle innovation.
Instead, in an AI world data architects must account for the evolution to open-architecture systems built for interoperability and scalability. This future lies in dynamic, entity-centric strategies that enable AI agents and automation to thrive.
Building Robust Entity-Centric Data Foundations
For investment groups to thrive, they must reimagine their data foundations. That means defining robust entity definition - the equivalent of infrastructure highways that enable different systems, data and AI agents to connect.
Technology like Jacobi’s Data Engine helps investment groups build entity‑centric data foundations by unifying portfolios, models, clients, and analytics in an open, programmable Data Lakehouse—unlocking the governance and interoperability needed for AI Agents to deliver to their full potential.
The winners will be those who quickly master the entities that matter - and build data strategies that let AI do the rest.
AI Investment Management Insights
The Investment Industry Must Urgently Rethink its Target Operating Model in the Age of AI
