Lots of Data, No Standards Equals Little Value
September 22, 2010 – Financial firms are centralizing the management of their data. And are willing to pay extra to make sure the data is accurate.
That sounds like pretty good news: Now comes the bad part. Financial firms still haven’t gotten their acts together when it comes to standardizing data terminology and that’s making data integration a lot harder. That is according to a 2010 benchmarking study just released by the Enterprise Data Management Council, which worked with financial service consultancies Headstrong, Element22 and Lepus.
The report is based on an analysis of 61 financial firms worldwide. Of those, six were universal banks, 34 were fund managers, 21 were broker-dealers.
Worse: Given the state of data management, firms may not be prepared to meet the requirements of a new Office of Financial Research now envisioned under the financial reform legislation. This was not covered in the report.
The goal of the OFR, housed within the Department of Treasury, will be to collect data from large financial institutions so it can analyze systemic risk – the potential for the financial demise of one firm to affect others in the global economy.
The government already has plenty of data in its fingertips. So do financial firms. But without data standardization – uniform data attributes, identification codes and data models – it will be pretty hard for firms to come up with the necessary information for the OFR to evaluate and it will be hard for the OFR to make meaningful comparisons on the financial health of different firms.
“No universal bank or sell-side firm claims any degree of standardization either enterprise wide or across individual business units,” said the report distributed at an event hosted by the EDM Council and consultancies on September 15. “From a collective industry perspective we are still quite far away from meeting standards objectives.”
Buy-side firms are furthest along the standardization path with 43 percent of top-tier and 39 percent of tier two and tier-three either standardizing data terms and attributes across enterprise-wide or individual business units.
The lack of data standardization is also making it far more difficult for firms to integrate data across multiple applications and have a consistent view of that data. Others are data quality, the inability to commission legacy applications, and lack of resources, funding or business sponsorship. That means nobody is willing to take charge of the data integration project.
According to the survey, Tier 2 and Tier 3-sell side firms cited lack of resources and funding as obstacles to data integration while Tier 1 firms cited data quality. For buy-side firms funding was also the biggest challenge for Tier 1 firms while data quality is a bigger challenges for Tier 2 and Tier 3 firms.
So where are firms spending their budgets for data management? Acquiring raw data and data feeds accounts for the highest percentage – 23 percent. But data cleansing and reconciliation aren’t far behind at 19 percent. Data integration amounts to 16 percent and data governance is 7 percent.
Maybe if there were data standards firms wouldn’t have to spend so much money on data cleansing and reconciliation or data integration. The study showed that one-third of the total number of firms surveyed are manually scrubbing more than 20 percent of their reference data every month. That figure jumps to 60 percent of sell-side firms.
What’s left to do? For one, the OFR could devise the data standards. Or bar codes could be an answer.
In releasing the survey, Michael Atkin, managing director of the EDM Council, came up with his wish list for the OFR to standardize: business entity identification codes, identification codes for over-the-counter derivatives and other financial instruments for which there are no identifiers and a data dictionary of how to define data attributes. The EDM Council itself has been working on
Nobody wants a government agency run by bureaucrats to dictate what the standards should be? Hopefully, the creation of the OFR itself could give the financial industry the impetus to work together and come up with standards of their own.
And in this sense there is some light at the end of the tunnel: “The overall level of standardization is low, but a majority of firms are in the process of addressing this gap,’’ says Atkin.
"Firms recognize the nature of data content standards for internal alignment and data comparability. They understand the EDM integration and the value of both architectural and semantic standards and want them implemented as soon as possible."
This story originally appeared on Securities Technology Monitor.