If you peruse the headlines about the global economy, you'll see it's replete with quotations from the "chief economists" of the world's premier financial services institutions. The chief economist role is a critical one for the prediction of trends, analysis of current conditions and to capture lessons from the past -- in addition to providing "look ahead" guidance for a company's strategy.
As we know, however, investment strategy is only part of the game. Technology plays a vital part in every financial company's operation. With IT intensity (the demand for computation cycles, storage and its related expense) rising faster than the GDP of any nation and at rates that exceed any firm's revenue growth, the title of CIO could just as easily be chief technology economist, or CTE.
Plus, if you consider that the most rapid rise in technology usage and expense is outside of IT (the non-IT technology such as big data in marketing), there is a need for foresight and insight into technology economics -- from both "input" and "outcome" perspectives. Hence, there is even more of a need for companies to get control of their technology economics. In fact, this may even be more of a "Fed" role that sets an organization's technology "monetary policy" and covers everything from "run the business" to outright innovation.
The Technology Economy
World-class CIOs are already well aware of this trend, though they may not define their actions in such stark terms. But they do seem to consider the steady state of the company's technology economy and the core components of technology expense.
Technology expense certainly has a commodity component. Tech commodities are evident in common capabilities and resources, such as supplying and operating infrastructure resources -- servers, storage, mainframe processing, networks, desktops/ laptops, mobile devices. Perhaps various categories of labor and external services could also be viewed as commodities. The importance of separating out commodities is that they can readily be "marked to market," so a financial services organization can see exactly how much it pays for a commodity compared with its peers. The market, in this case, is the technology marketplace with service levels normalized to enable a true apple-to-apple, or commodity-to-commodity, comparison.
Technology expense also has a demand component. Demand has to do with the volume of IT resources an organization has relative to the size and nature of its businesses. For instance, does a firm have too much available storage capacity for an organization its size?
However, can a company mark to market a demand? While demand might seem to be difficult to benchmark, a company can obtain useful information by comparing its level of demand and be able to answer the question as to why it has more (or fewer) resources relative to business size and nature versus its peers. In addition, demand growth is key to understanding a number of trends. For example, let's say one company's number of SQL databases grew 179% at a time when the firm's revenue was declining. Or why is a firm's expense on applications maintenance more than double that of peers? Does the organization have too many applications, or is the company neglecting application rationalization?
In the world of application development, similar demand questions can be addressed by benchmarking. On a product-by-product basis, is a financial IT organization investing more (or less) in new systems than competitors? There could be very sound answers to these questions. For instance, perhaps the answer to the growth in SQL databases is that more databases were created to support big data analytics. But these demand questions need to be asked, since 50% of technology costs are demand driven.
[Check out this session: IT Is Your Business, Run It Like One at Interop, which runs from September 30 through October 4 in NYC.]
In addition to the commodity and demand components of technology expense, every organization has a "cost of doing business the company way" component. While this component doesn't exactly roll off the tongue, almost everyone knows what this is (the quirks, or culture, that every company has). This is where unique company characteristics and technology service levels impact technology economics. For example, one investment bank maintains network bandwidth to enable a 7x market peak so there are no constraints on trading. That is "its way" and costs the company (along with some other unique business aspects) almost $500 million more per year (15 cents a share) than competitors.
When you look at commodity, demand and cost of doing business, on average for CIOs/CTEs who applied this model, only about 30% of infrastructure optimization benefits were related to commodity platform unit cost reduction. Demand (41%) and the "cost of doing business" (29%) accounted for 70%.
So how do CIOs or CTEs actually accomplish benchmarking? Here are some characteristics common across the CIOs who act like CTEs in their use of benchmarking and its application to the technology economy:
• Commodity benchmarking: This is a fairly straightforward process and considers unit cost and scale versus peer companies and even technology suppliers (e.g., what does it cost Amazon to run a server?).
• Demand benchmarking: For infrastructure benchmarking, this can be done by considering key ratios of infrastructure capacity to business parameters (e.g., revenue, transaction volume, head count).
• Application maintenance benchmarking: This can be done by assessing application change rates and percentage of time and expense for "fixes" and such.
• Application development benchmarking: For application development, this can be done by calibrating absolute spend on new development on a business product basis with consideration to development cost drivers (labor rates, productivity, needed quality, etc.) against peers over a multiyear period.
• Company culture benchmarking: This is more complex but involves stripping away exceptional service levels not evident with peers and determining the associated incremental cost. It's easiest to do this from the "top down" -- attributing a top-line gap to this phenomenon and then drilling down.
However, beyond these benchmarking basics, the CIO/CTE knows that technology success is measured in results, or outcomes. Technology outcomes should link to business results, such as revenue growth, revenue protection, risk management, and cost avoidance and reduction. Moreover, the CIO/CTE also knows that these outcomes are related to operational efficiency/excellence, enhanced customer intimacy/relationships and product leadership/innovation.
Ultimately, to manage the technology economy from an outcome perspective, key performance indicators need to be defined to chart the outcomes of technology investments -- both expected and actual. Technology investment without such KPIs is essentially a high-risk unguided (and expensive) missile.
CIOs/CTEs need benchmarking and the proper KPIs because they are dealing with issues that will govern and define an organization's technology economy far into the future, and will be the underpinnings of a true competitive advantage. These issues are rarely addressed head-on today but are likely the determinants of enterprise success in the future.
Dr. Howard A. Rubin is a Professor Emeritus of computer science at Hunter College of the City University of New York, a MIT CISR Research Affiliate, a Gartner senior adviser and a former Nolan Norton Research Fellow. He is the founder and CEO of Rubin Worldwide.