The heat is mounting on the Windows 2000 rollout. Despite Microsoft's bullish comments about Win2K's pace of acceptance during the company's conference call with analysts after the release of its quarterly earnings, and despite the long-awaited release of higher-end, Pentium 4-based computers, many observers wonder whether Win2K adoption will fall short of expectations. They argue that with a widespread economic slowdown gaining steam, companies will have to cut back on capital expenditures, including IT investments. Moreover, if consumers are a harbinger of broader spending trends, the post-PC era has begun. Corporate focus could be shifting to IT investments that can open up corporate networks to a wide range of Internet-access devices ranging from smart cell phones to personal digital assistants (PDAs).

If a faltering economy does lead to a slowdown in overall IT spending, the Win2K migration rate could be hurt, according to Research UPDATE's exclusive analysis of data provided by Survey.com. Graphs 1, 2, and 3 reveal what IT managers who haven't yet begun to evaluate Win2K but plan to do so in the future believe to be its worst features. Notice that more than 50 percent of those managers put cost first. This response didn't vary whether managers anticipated evaluating Win2K for implementation on servers, desktops, or laptops.

In fact, unlike other aspects of the Win2K acceptance process in which the platform on which managers planned to implement the OS made a difference in their feature evaluation, the results in this case are remarkably consistent. Not only is cost by far the leading concern; each group of managers mentioned the same set of features in the same order.

While this result is unusual, perhaps we should have anticipated it. Corporate decision-making often follows a fairly predictable course. Concerns about cost, interoperability, reliability, support needs, and performance are enough to delay a purchasing decision.

But are those concerns well founded? When we explore this question, a mixed picture emerges. Graphs 4, 5, and 6 show the results for a group of IT managers who have completed their deployment of Win2K. The bad news is that, once again, total cost tops the list of worst features mentioned. The good news is that after Win2K is deployed, its cost diminishes in importance. Less than half of those who have deployed Win2K cite cost as a negative factor, a drop of about 10 percent.

The data reveals other differences in perception between managers who have passed on evaluating Win2K for the time being and those early adopters who have already completed their deployments. Those who have deployed Win2K are much less concerned about reliability and stability issues than those who haven't started the evaluation—in fact, reliability and stability don't even make the top five list for the early adopters. On the other hand, those who have deployed the OS find management issues, including storage management on servers and remote management, more troubling.

We can draw several conclusions from these results. IT managers are worried about Win2K deployment costs; if general IT spending slows, the rollout will slow down as well. But concerns about system reliability by managers who haven't started the evaluation process are, to a degree, overblown. IT managers should focus instead on the management and support issues involved with the migration.