It is often said — particularly by those who engage in such behavior — that rules are made to be broken. Sometimes recklessness is rewarded, but more often it’s fatal.
Most of us, for example, would never intentionally run a red light. But a great many of us have rolled through the occasional stop sign. We look both ways, honoring the spirit of the law if not the letter, and we keep moving.
There’s a corollary in business: rules are vital but also constraining. Companies have good reason, for example, to insist that IT investments meet some sort of ROI threshold. Capital spending is under intense pressure, and the haste associated with Y2K and Euro conversions, together with the rush to E-everything, caused both a spending glut and substantial disappointment in the results of major projects.
But are we overcompensating? Earlier this year, Lands’ End CFO Don Hughes addressed a roomful of CFOs and happily admitted that his company had spent a substantial sum to add a “gee-whiz” feature to its Web site without worrying much about a rigorous ROI analysis. The feature, dubbed My Virtual Model, allows consumers to enter their personal measurements and “try on” clothes in a sort of virtual dressing room. The company also captures those measurements and uses them to manufacture custom-tailored slacks.
The major reason that Lands’ End didn’t conduct a formal ROI review was that there was nothing to compare the project to. But it appeared to be both feasible and valuable; with online sales now accounting for about 20 percent of the company’s revenue and growing three times faster than the mail-order business, Hughes has no regrets.
Business-school professors Jeanne Ross and Cynthia Beath applaud such thinking and caution companies against turning ROI into a security blanket. In interviewing executives at 30 large corporations, they determined that some new thinking is called for. IT investments, they argue, should be broken into four distinct areas: transformation, renewal, process improvement, and experimentation.
Transformative projects are those that improve core infrastructure, such as an ERP rollout or expanded Web-site capabilities. For these projects, funding should be based on executive-level allocation, not a hard-and-fast business case. The same holds true for experimentation (using IT to test whether customers can self-serve, for example).
The business case still matters for renewal projects (those that reduce the cost or boost the quality of IT services) and particularly for process-improvement efforts (which have a specific operational focus and so presumably can be quantified with some accuracy), but that’s not the same thing as a blanket reliance on an ROI formula.
Moving to such a system is not easy, but Ross, principal research scientist at the Center for Information Systems Research at MIT’s Sloan School of Management, points to UPS and others as examples of companies that understand that a more-sophisticated view of IT investment is essential. “Some companies have become more conservative in their IT investment approach,” she says, “and that is usually synonymous with a focus on ROI. But such companies don’t tend to be categorized as world-class.” Ross agrees that it can be tough to distinguish between the types of IT investments she describes, but she and Beath do offer some help (see their paper “Beyond the Business Case: Strategic IT Management.”)
By focusing on individual projects that are either approved or rejected based on their payback potential, Ross says that companies miss the chance for “happy surprises,” or new ways to use data and IT infrastructure that allow them to stay one step ahead of the competition.
CFOs might counter that getting verifiable payback to a project is itself a happy surprise, and no doubt the business case is not to be easily dismissed. But it should not codified into law, not if you want to keep rolling.