Information is the lifeblood of investment management. What strikes me as peculiar about this statement is that while most investment managers would agree with it, the reality is the industry's approach to the management of data is only just receiving the attention it deserves. Put another way, I think data in investment management is a little bit like the water we drink every day: none of us can function without it, we are not always sure where it comes from, some of us have no idea how much it really costs, we don't mind if we waste it, we don't like sharing it, and usually, we only find if it was bad for us after we have drunk it.
Data management has always been around in some form in investment management; however it started to gain some momentum as a recognized discipline for transaction processing in the back office over 10 years ago. Unfortunately, that is where it has been stuck, both technically and culturally. If ever things needed to change then it is now, given the volatility of markets, the pressures to control costs, client demands for greater transparency and the flood of new regulation.
So how could we change things and make data management more of a foundation to be built upon rather than being perceived as a secondary function housed within the back office? I have heard a lot of talk about data governance at many of the conferences I attend, but try selling that as a concept to analysts, portfolio managers and trading staff.
Governance and policies are vital to parameterize how data should be used and managed, but unless the "stick" of data governance is supported by the "carrot" of "what's in it for me," you will always be fighting an uphill battle against human nature.
Looking at the "carrot" side of the equation, one of the biggest problems with both vendor and internal systems is the lack of focus on the human interaction with the data. Systems are designed to work in the background, out of sight and out of mind, feeding downstream systems with the data they need.
What's the problem with that you might say? Well not much if you confine yourself to traditional views of what data management is used for, but a lot if you are wanting to make the most of the data and information your business is built upon.
Firstly, why would you choose to exclude front-office staff from being involved in the data management process, relegating them to telling you that the data is "wrong" only after it has been distributed out to the systems they use? Front-office staff often know more about the quality and "correctness" of the data than back-office or technology staff, so why not leverage this knowledge through providing them with the tools and means to contribute to and benefit from access to the data.
Secondly, through getting business users to be able to interact with and use the core data systems, not only would data quality improve but front-office staff would be less tempted to create their own "tactical" spreadsheet and database solutions – "tactical" solutions that soon become a "permanent" fixture in many business processes and a major source of cost and operational risk.
Finally, business staff do not care what type of data they are dealing with, they simply know what data they need to carry out their business. In this regard, data management systems should provide easy access to data across all asset classes, but also across all types of data (reference, static, market, derived) and temporalities (static, semi-static, historic, intra-day and real-time).
With relationships between data and across asset classes becoming ever more intertwined, providing unified access to all the data needed by the business is becoming increasingly more important to reduce costs, reduce risk, gain new business and be able to respond to regulatory reporting requirements both now and in the future. So get the business plumbing right, and let the information flow.
Brian Sentance is the CEO and one of the founding directors of Xenomorph, an analytics and data management solutions provider to global financial institutions.