Making assets actionable: the DAM approach to alternative data in investment research

As a supplement to conventional sources of information on prospective investments, such as company reports and financial statements, ‘alternative data’ have an increasingly important role for investors seeking the elusive ‘alpha’. From weather forecasts to consumer spending, satellite images and online reviews, a vast wealth of often unstructured data can provide valuable clues to the future performance of companies and sectors.

For suppliers of research to investors, the ability to incorporate this kind of data into their reporting is key to maximizing the value to their research and making it stand out from more conventional offerings.

Unstructured and incompatible?

But, compared with the highly structured data obtained from traditional sources, this kind of unstructured content presents several challenges to ingestion and utilization within a conventional research environment:

• Compatibility. Alternative data may include a variety of file types, including images, video and proprietary formats, as well as normal numerical or tabular data. For conventional systems this makes ‘onboarding’ and storage of the data problematic.
• Findability. To be useful, the data assets must be ‘findable’ which demands careful metadata tagging. Without this, even the most valuable assets add nothing to the reporting.
• Accessibility. Busy analysts will not make use of data unless it is easily visible and accessible within their workspace.
• Visualization. To be useful and comprehensible to the end-user, the data must be presented in an attractive and accessible way - going beyond the usual text-and-table approach.

The DAM approach

These challenges are familiar to those working in the news-media and allied sectors, where, over the last two decades, the need to turn unstructured multi-format digital assets into value-added news products has been a pressing requirement.

This has driven the development of powerful digital asset management (DAM) functionalities capable of ingesting large volumes of digital content in multiple formats and making it immediately available to authors and designers engaged in creating content for shorter and shorter news cycles. AI and ML tools are increasingly being used to automate the routine processes (such as metadata tagging) involved in the management of this kind of content.

Applied to the creation of investment research, these DAM capabilities offer a high-performance solution for the management of the entire spectrum of data types, both conventional and alternative, maximizing their contribution to the output of authors and analysts.