What data is needed to find opportunities in financial research

Panels at the Bloomberg Enterprise Tech & Data Summit said firms are combining traditional indicators with alternative sources—like consumer transactions, foot traffic and digital news—and using AI to turn unstructured data into usable signals. These changes are widening who can access and act on data inside firms and are bringing discretionary and systematic approaches closer together.

Panelists at the Bloomberg Enterprise Tech & Data Summit described how firms are expanding the types of data they use and the tools that process it. State Street, for example, now produces a continuous inflation dataset that blends traditional indicators with alternative sources such as observed consumer spending and digital news. Speakers noted that vendors provide a range of alternative data products and that AI and engineering work are central to converting unstructured information into signals. What we know: - State Street combines traditional indicators with alternative data (consumer spending and digital news) to produce a continuous stream of inflation information for clients. - Bloomberg offers alternative data via Terminal and data feeds, including consumer transaction analytics, foot-traffic data and web-traffic analytics. - Discretionary teams are able to examine a broader universe of securities because scalable data infrastructure makes more names and signals accessible. - Quant teams are increasingly translating unstructured data into structured signals for models, supporting closer alignment between discretionary and systematic styles. - Experts highlighted agentic AI as enabling tool-calling and workflow automation, and firms stressed the need to shape and architect data for AI consumption. - Several speakers emphasized making data accessible across employee levels and providing low-code or no-code tools so more staff can test and analyze data. Summary: Panels described that richer, blended datasets and AI-enabled workflows are changing how firms generate and scale insights, leading to greater overlap between discretionary and systematic approaches. These developments depend on data architecture, engineering effort and broader access to data within organizations. Undetermined at this time.