Likely all financial services organizations are now moving into their 2013 budget planning cycle while doing contingency planning for the rest of 2012. However, few if any are "stress testing" their technology economics.
An IT Stress Test is a scenario-driven analysis used to assess the ability of the IT organization to scale down (or up) its costs, personnel, capacity and more based on business drivers and conditions, such as global economy changes, revenue changes, regulatory changes and changes to core business transaction volumes. It helps an organization calibrate another form of risk -- the risk of inflexible IT economics impacting the company's bottom line performance.
[Chief Data Officer's Role Growing in Importance.]
With the typical IT organization having a fixed cost base of 65 percent to 85 percent of IT expense, and with capacity needs growing independently of revenue, it is critical that IT leadership (and business management) have an understanding of how technology economics will impact the bottom line in a range of market conditions. And market conditions certainly have shown enormous variation over the past five-plus years.
Making the Grade
The first step in preparing for an IT Stress Test is to create a clear view of the organization's fixed and variable costs. In the simplest of models, fixed costs are defined as those expenses that cannot be impacted by changes in the annual budget cycle, while variable costs are those than can be influenced in a similar time period.
However, you need to be aware that such definitions are "fuzzy" in that something that is variable in the beginning of the year (e.g., labor costs, which can be influenced by downsizing) can become fixed later in the year -- look to downsize in November or December, for example, and the economics likely will act as if they were fixed costs until the end of the year.
From an accounting vantage point, expenses that are fixed typically include depreciation; real estate; long-term, volume-independent licenses; outsourcing deals; and similar costs. As noted, labor costs (from contractors' fees to employee compensation) usually are the first to come to mind as variable. But care must be taken on the associated labor "kit" (equipment, connectivity, space, administrative overhead, etc.) which often is fixed. Other sources of variable expenses may relate to on-demand capacity, resources with costs based on consumption and so on.
One way to generate a fixed-versus-variable-costs profile is to use a "fuzzy set" model: Characterize expenses on a scale from 1 (fully fixed) to 5 (fully variable), and then compute a weighted average for the total expense base. Next, construct a model with a focus on downward compressibility -- what is the floor for bringing down all variable expenses? -- and upward scalability -- how do costs move upward with increases in business (e.g., revenue, transactions, head count, customers, assets, etc.)? Then use historical data to assess how your expense base will perform under the variations your business has experienced. Finally, build scenarios looking forward -- both upside and downside. This is the "IT Stress Test."
In assessing the impact of such "stress," watch the key IT ratios of expense versus revenue and expense versus operating expense. But also take the analysis a step further: What is the influence of IT expense relative to earnings per share? What is the influence of IT relative to return on equity? Simultaneously, though, be sure to consider the influence of under-spending on IT. Watch the correlation of IT intensity to gross margin as an indicator of levels of underinvestment.
It is not likely that your organization will be able to produce a comprehensive model of its technology economics. The process of constructing the stress test itself, however, will lead your organization to collect needed data and create hypotheses about the dynamics of IT and the business. In fact, it is likely that in attempting to do the analyses sketched out above, you will be taking yet another step toward charting your organization's technology economics and will be well on the road to becoming a master of them instead of a victim.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 Advisor, and a former Nolan Norton Research Fellow. He is the founder and CEO of Rubin Worldwide. Dr. Rubin is ... View Full Bio