Comment: Can firms trust ROI forecasts?

Times are hard for network managers. Shrinking or frozen IT budgets reduced the scope for system upgrades in the past year, leaving managers scratching their heads about how best to eke the last drop of performance out of the hardware and software they already have.

And while it is anyone's guess whether the purse strings are likely to be loosened in 2003, it seems that financial caution will remain for the first half of the year at least.

Which means IT managers will have to fight that much harder to secure any much-needed investment in LAN or WAN infrastructure. More specifically, they will have to convince the board that an upgrade is absolutely essential to ensure that their firm remains competitive. Or they will have to show the chief financial officer that any new equipment will pay for itself inside a set period of time. In most cases, IT managers will probably have to do both.

Matching a specific upgrade against a network problem that needs to solved, or against a new application or business process that needs to be supported, is by far the easiest part of the task. The complications, ambiguity and uncertainty, all derive from the return on investment (ROI) half of the equation.

Because no matter how hard some vendors try to convince us otherwise, accurately calculating ROI figures for individual companies is a very difficult, if not impossible, thing to do in most cases.

That is not to say it cannot ever be done. But the ubiquitous ROI calculator, featured in the travelling salesman's armoury and on Web sites of firms selling anything from voice over IP (VoIP) to packet monitoring software, is not the way to do it. In every case, these are simply poorly disguised marketing tools that arrive at the same conclusion - namely that the kit you buy pays for itself within a year - no matter what figures are entered by prospective customers.

The simple truth is that ROI depends so much on individual circumstances that only a detailed analysis of the situation in each case can give anything near an accurate prediction of how much a firm can save by opting for one technology rather than another.

And the only way specific circumstances can be properly assessed and quantified is through an independent examination of existing systems, the applications and transactions that those systems support, and by developing an understanding of how individual departments and employees use network resources.

Some vendors will undertake to perform this type of detailed analysis and produce a meaningful report based on their findings, not least because it least provides them with a foot in the door of any potential sale.

Commissioning two or more competing vendors to conduct these preliminary examinations will provide a good cost comparison and a certain measure of reassurance if they all come up with roughly the same conclusions.

While no one can be absolutely certain that upgrades will ever pay for themselves, the ability to distinguish intelligent, fact-based forecasts from wild calculations and unsubstantiated predictions is key to ensuring that scarce money is not wasted.

Have your say: reply to IT Week

More IT Week Comments