Making the Whole Solution Greater than the Sum of its Parts
Much of which was custom built and has been patched and upgraded over the years. Over time, new technological acquisitions have only exacerbated the situation. Weaving new systems into the stack and seamlessly exchanging data between the entire company, not to mention external users, is the equivalent of a technological game of Jenga.
In the end, we rely on clunky workarounds and manual solutions to get the job done, when a total system overhaul is needed. However, a technological restructuring could take years and millions of dollars to complete. Many companies ask themselves, "When do we put our toe in the water?" "What is the right technology and when is it worth the investment?" "How do we put all these pieces together to get the most out of it?" It can be expensive and painful to implement. It's even harder if you are not a big firm with the budget to undertake such a project.
Oftentimes, the workarounds are "good enough" and the upgrades so daunting that organizations just chug along as best they can…until they simply can’t and their shortcomings become painfully obvious.
There simply has to be a better way. How can we still use our legacy technology, yet incorporate new FinTech services, and not have to piecemeal the whole thing together? The answer, in my opinion, can be found through leveraging the power of semantic web ontologies.
How can we still use our legacy technology, yet efficiently and flexibly incorporate new FinTech services?
My team at UCC has been working with David Saul, Chief Scientist, and other State Street experts on a proof of concept that took a sliver of data, using bonds and equity markets, and mapping it against FIBO (Financial Industry Business Ontology). FIBO, with the Object Management Group (OMG), defines data meanings in our industry. With our approach, all of your data (both old and new, structured and unstructured) stays in your proprietary systems, databases, Excel files etc, as it stands. This new ontological layer sits on top of this technology stack and extracts information based on characteristics to meet business requirements. The approach essentially creates a map of your data, no matter where it is stored or in what format.
Since this is a filter on top of pre-existing systems, data can remain in any system or database. The data is extracted and filtered through the ontology layer, dramatically reducing the manual work of searching/pulling/correlating/cleaning/sorting it all. This approach has significant business value. After all, when the next generation of FinTech technology comes along, you won’t have to replace your entire backend setup or undertake expensive system migration projects. You will just add it to the mix of tech in play and put the ontology layer on top.
Many financial services institutions are grappling with adopting the right technology. This ontology layer provides you with the ability to use multiple distributed systems and pull data from anywhere. Instead of putting everything into one big bucket, leave the buckets where they are and just put this layer on top.
In a recent SEC announcement regarding tagged data, Commissioner Kara M. Stein said, "Today, the Commission is proposing a single reporting format for public company reports that is both human and machine readable. Computers can quickly parse the data and humans can view the document in an ordinary way." As Saul has mentioned, "I don’t believe I could have said it better myself. To make data both human AND machine readable is exactly what we are trying to do. That is at the heart of what I have been advocating for through the years."
While this ontological system is still in its early stages, we’ve had promising results with our proof of concept. The next step is to test our system with more data from a different sector and see if we can replicate the results. We want to make sure we are doing and we are applying it. Tech for tech’s sake just adds to the mess. Our goal is to take the pieces and build something that fits seamlessly together.
Philip O'Reilly is Senior Lecturer in Information Systems and Co-Director of the State Street Advanced Technology Centre at Cork University Business School, UCC. Phil is a fan of gadgets, which rarely reach "legacy status" thanks to his two small girls.