Why Reference Data Distribution is Critical to Successful EDM
John Randles, CEO PolarLake
There has been a massive surge in the amount of data fund managers have to deal with. This is a result of globalization, the massive increase in trade volumes, more complex asset classes and the mix in asset classes, among other things.
As data volumes have increased so has the burden on the financial markets industry to get the right data to the right systems at the right time. This has remained a constant, if increasingly tricky, challenge.
The emergence of Enterprise Data Management (EDM) in the last few years has made great strides in the move to tame the torrent of reference data flooding the industry. The key effort surrounding this has been on centralization, in an attempt to capture and cleanse data before making it available to fund managers.
Typical data management projects spend 44% of their time on the data acquisition process, 20% on scrubbing and 36% on reference data distribution, according to research by the Tower Group.
This is significant because the primary concentration of the industry when it comes to EDM has been on data capture and scrubbing (64%). Typical EDM projects build all the infrastructure and functionality in order to create the Golden Copy. This focus means the efficiency and effectiveness of data distribution often gets overlooked, or plays a secondary role.
The average number of systems within financial markets firms containing client and counterparty reference data is 44. The average number of systems containing instrument reference data is 37. So when you achieve a Golden Copy the job is still not done. Getting the reference data to the fund manager can still be a momentous task.
The scale and number of the systems involved means to modify each system in order to access reference data is too big and onerous a task for anyone to take on. The combination of home group, commercial, unsupported commercial and obscure systems, means organizations are often afraid to interfere with them because they work (even though not to maximum efficiency) and are mission critical.
So, regardless of the investment made in the centralization of reference data, the ROI promises of EDM can only be delivered when the centralized is de-centralized and the myriad of downstream systems benefit from the Golden Copy.
Key to the value in EDM is the mitigation of operational risk, which can only be delivered and measured based on the quality of data in the downstream systems. Improved STP rates can only be delivered when the OMS, Portfolio Management System, Risk Systems, etc, all have the same quality data as the centralized Golden Copy.
This might sound obvious but in the rush simply to capture and cleanse data to produce Golden Copy it can often be overlooked. It is worth noting that 51% of all failed trades fail because of inaccurate or incomplete reference data or disparities between reference data.
Therefore the level of ROI and risk associated with EDM projects will always depend not solely on the production of Golden Copy, but on an organization's ability to distribute and synchronise reference data to multiple and highly complex downstream systems.
To see how this can be achieved, view PolarLake’s whitepaper entitled How can you leverage your Reference Data to reduce operational risk and costs?
2010