M Science, a leading provider of data-driven investment research and analytics, today announced the launch of its Unified Data Model and Model Context Protocol (MCP) Server, creating a modern data and AI infrastructure layer for institutional investors.
Together, the Unified Data Model and MCP Server are designed to help clients move faster from data ingestion to insight generation by standardizing access to M Science’s Analyst-Curated Data and enabling direct, programmatic use of M Science research and curated datafeeds inside client AI applications, internal copilots, and automated investment workflows.
The Unified Data Model provides a single, standardized framework for M Science’s Analyst-Curated Data, simplifying the ingestion and analysis of historically disparate data feeds. Built on a scalable star-schema architecture and enhanced with Change Data Capture, the Unified Data Model supports precise point-in-time analysis, historical comparisons, and robust back-testing.
The Unified Data Model is backed by the breadth and depth of M Science’s proprietary data ecosystem, which includes more than 1,440 key performance indicators and daily or weekly data on over 1,400 companies, with datasets updated in near real time. This coverage is supported by a rich historical archive, enabling longitudinal analysis across market cycles and more granular, point-in-time views into company and sector performance.
Complementing the Unified Data Model, the new MCP Server will give clients secure, programmatic access to M Science research and data through tool-based interfaces designed for modern AI systems. The MCP Server extends the foundational infrastructure that powers Maddie, M Science’s AI copilot, providing institutional clients with a flexible, programmable interface to M Science’s research and data intelligence, enabling deeper integration into proprietary AI systems and workflows. This includes integration with widely used large language model environments such as OpenAI’s ChatGPT and Anthropic’s Claude, as well as internally developed AI systems.
“M Science is focused on delivering not just differentiated data, but the infrastructure clients need to operationalize it at scale,” said Michael Marrale, CEO of M Science. “The Unified Data Model reduces friction in data ingestion and analysis, while the MCP Server will allow clients to bring M Science’s data and research directly into the AI-driven workflows they are already building.”
A defining feature of M Science’s platform is the connection between structured data and deep analyst context. M Science’s Analyst-Curated Data feeds, originally launched in 2018, have been continuously refined and expanded through ongoing enhancements in methodology, coverage, and validation. Combined with the firm’s extensive archive of historical and current written research, this creates a contextual intelligence layer that differentiates M Science from traditional data providers.
By leveraging MCP, clients will be able to access not only structured datasets, but also the research context behind them, helping create more explainable, auditable, and actionable AI-driven investment workflows.
The Unified Data Model is supported by flexible delivery options, including Snowflake Share, Databricks Delta Sharing, S3, API, and an enhanced user interface within the M Science Portal. The new UI allows users to explore data feeds in a single view, configure scheduled deliveries, and streamline discovery and access across datasets.
At the same time, the MCP Server supports a more flexible and scalable way to interact with M Science content, moving beyond static data delivery and traditional interfaces toward fully integrated, agentic AI environments.
“With the introduction of MCP, we’re extending M Science beyond our platform,” said Spenser Marshall CIO at M Science. “Clients will be able to access our research and Analyst-Curated Data in a programmatic, controlled way that aligns with how modern AI systems operate. The combination of structured data and deep contextual research is what makes our platform uniquely powerful in an AI-driven world.”
Key benefits of the combined UDM and MCP Server launch include:
- Unified Data Architecture: A consistent schema that simplifies ingestion, reduces data engineering overhead, and accelerates time to insight.
- AI-Ready Integration: MCP-based access that enables M Science data and research to be embedded directly into client AI systems, copilots, and agentic workflows.
- Point-in-Time Analytics: CDC-enabled architecture that supports accurate historical comparisons, backtesting, and longitudinal analysis.
- Workflow Efficiency: Standardized data structures and programmatic access that reduce operational complexity across data pipelines, research processes, and AI applications.
- Contextual Intelligence: Integration of structured datasets with M Science’s deep archive of analyst research for richer, more explainable insights.
- Flexible Delivery: Access through API, cloud shares, S3, the M Science Portal, and MCP tools to support a wide range of client infrastructure needs.
“The combination of standardized data and programmatic AI access represents a meaningful shift in how clients can use M Science,” said Marshall. “We’re enabling them to move faster — from ingestion, to analysis, to decision-making — while maintaining the controls, transparency, and context they require.”
About M Science
M Science, a Jefferies company, is a leading provider of data-driven research and analytics, offering differentiated insights derived from a variety of alternative and traditional data sources. The firm combines proprietary datasets, advanced analytics, and deep industry expertise to help institutional investors make more informed decisions.
For more information, please visit www.mscience.com or contact research@mscience.com.
View source version on businesswire.com: https://www.businesswire.com/news/home/20260602728905/en/
Media gallery
