Data management costs are always a concern to financial firm management teams. The main goal of these managers is creating a solution that reduces the overall cost of managing data, improves the efficiency of managing data and still maintains, or even improves, its quality. Well-meaning managers attempt to quantify the costs and create plans to boost efficiency but are often stymied by the reality that most firms seem to have relatively little fat in their data management process. When making the case for improving the data management process through investment in a data management solution, these managers find it hard to justify the costs of development or outsourcing against the savings that will be generated.
When analyzing the cost of data management in financial firms, the discussion typically centers on the following three areas:
One of the jobs of a data manager is to contain the costs associated with these areas, while still delivering the data required by the business in order to deliver services to their clients. In most cases, the end result is that reduction in cost is synonymous with a reduction in staff or data feeds.
Understanding that this is true, senior managers in these firms look to the data managers to work with IT to improve efficiencies in order to either reduce staff, reduce the need to increase staff or reduce the number of feeds taken in. Unfortunately, the number of feeds is usually an outgrowth of the data quality of the feeds and the trade off from reducing feeds often results in the need to hire more staff to deal with data quality inquiries. This is an unacceptable outcome in many of these firms. This means the data managers lean more toward reduction in staff.
Data management staffs at many financial firms are rather small and the impact of a reduction in staff is too small to make a dent in the overall data management budget. Even when there is large enough staff to reduce, rarely does the savings from those reductions justify the expense of the projects needed to accomplish the savings. What most managers fail to notice is that many of the costs of data management are hidden within the business. Tackling these hidden costs will result in significant savings without impacting the business. In most cases addressing these hidden costs will actually enhance the business. The most egregious of the hidden costs is what can be termd the “virtual data analyst staff.”
What is a virtual data analyst? Virtual is defined by Webster as “being such in essence or effect though not formally recognized or admitted”. This definition works perfectly to describe the phenomena we are talking about here. A virtual data analyst is one that does the job of a data analyst but is “not formally recognized or admitted” to be one. As we will come to see, these virtual staff members do two things, they create a “hidden cost” for data management, one that is not accounted for when considering the budget for security data and a “hidden cost” for these downstream areas that must make up for reduced efficiency of the staff they have “lent” to data management.
Let’s look a little further into who these virtual data analysts are. As stated above, a virtual data analyst is any person in a firm, not designated as a part of the data management team, which performs work that is normally understood to be data management. Typically, the types of functions performed include, but are not limited to, verification of reference data such as bond coupon, maturity date, interest start date, call dates, etc., comparison of prices from various sources and meshing of corporate action data. Much of this work is a byproduct of other concerns such as downstream calculations accrued interest or allocation of cost. In many cases, the person doing this work may be a highly skilled and highly paid employee such as an accountant. They may be spending large amounts of time doing work that could be easily done by a lower paid technician or, if there was more confidence in the data, not at all. All of this ancillary work interferes with the work they are being paid to do and performing this work may even force additions to the staff to meet the deliverables of the downstream group.
The question becomes, then, how do we quantify an entity that we have already admitted does not physically exist? In one case, the data manager decided the only way to do so was to analyzed the work done by downstream areas. He contacted representational groups for the various global functions that consumed both reference and pricing data. He created a series of questions that captured the time that the non-data staff spent on what is essentially data management functions during a specific period. From these data, assuming it was representative of similar groups around the world, he was able to extrapolate a total number of person-hours globally dedicated to these functions. From this, it was easy to calculate the person-years, which was translated into the equivalent number of virtual staff. In addition, in order to validate the numbers drawn from interviews, he also quantified the number of inquiries sent to the dedicated data analysts. With these two pieces of information, even though individuals spent relatively little time on the functions, he was able to determine that the firm had a virtual staff 20 times the size of the dedicated data management staff.
The interesting outcome of this exercise is that once this virtual staff number is known, the argument to fund any data management solution is no longer a guessing game or a discussion of potential benefits to the business. Instead, the costs of the solution can be quantified by a reduction in virtual data analyst staff which then increases the actual staff, through increased capacity, of the downstream areas. Productivity, then, increases through the proper use of expensive resources and decreases the rate of hiring similar resources, to meet any new business needs. Virtual staff converts back into real staff. The benefits of the data management solution are now more tangible. The decrease of the virtual staff, which can be evidenced through service desk interaction, enables senior management to quantify the return on investment from the data management solution much more accurately.