Pushing back the alternative data frontier
Martijn Groot 2 minutes
Simply put, we have seen a rapid growth in data production channels (e.g. sensors, web crawling, geospatial information, regulatory push for pre/post trade transparency), new approaches to acquire and aggregate information (for instance through new data marketplaces and alternative data feeds) and, up to a point, new methods to knock it into shape for actionable use.
Data management so far often seemed like maintaining a walled garden: careful pruning and controlling a predefined data set for operations and compliance, in any case well defined use cases. What has changed has been the arrival of the wildflowers and the weeds of the alternative data jungle – as well as the opportunity that comes with that. Gardening needs to evolve…
Today’s alternative data can be tomorrow’s mainstream data and one way to define alternative data is to label it as ‘all data not (yet) commonly used by the financial services industry’. Due to new data marketplaces, new data sources and the incorporation of these new data sets into larger enterprise data products, the alternative data frontier has been pushed back. Developments in curation tools that can make alternative data actionable have sped up that process.
The importance of these curation tools speaks to one of the paradoxes in alternative data; to properly access, search, interrogate and integrate data into business processes you need to impose some structure. We could also refer to the whole category as ‘alternatively structured data’ rather than unstructured data. Quite literally, something without any structure is random noise. Curation can start with tagging images with geospatial coordinates and date/time, marking up newsfeeds and earning call transcripts and range to correlate vast amounts of data with entities, specific instruments and risk factors.
Where we are seeing the biggest developments is in fast closing the gap in the tools to integrate these data sources into day to day business workflow processes. Crudely put, if you can’t put a data source to use, interest in it will quickly fade. Adoption areas can range from compliance (early use cases were studying behavioral patterns in large transaction data sets) to risk assessments. The data intensity of risk and reporting processes is likely to continue to grow. However, additional data sets are also fast entering into trading and investment decisions. Their use in those processes will be another factor pushing back the alternative data frontier.