Few Projects Are Over Budget
|
If a penny saved is a penny earned, how many pennies are earned when a project is kept on track? One of the best ways to manage costs has got to be keeping projects on budget. So we asked IT executives what percent of their IT projects came in on, under or over budget in 2006. |
The numbers:
On budget 66%
Under budget 17%
Over budget 17%
N=256
These numbers do not account for scope increases -- so these numbers only show whether projects stayed on budget even after the scope was adjusted. That's no small matter, and were I to ask the question over again I'd also ask about scope. Still, it's remarkable that the over-budget figure is just 17 percent -- just one project in six.
I suspect that to many of you, this figure will seem not just surprisingly but impossibly low -- especially if you believe (as so many do) that two-thirds of IT projects fail. But that two-thirds figure is a myth. It's based on an erroneous read of the Standish Group's oft-cited "Chaos Report" surveys. As James H. Johnson, chairman of the Standish Group, told me when I investigated this for a short June 2006 story entitled "The Myth of Frequent Failure" "we never said two-thirds."
That story never made it to our Web site, so it bears repeating here. This is what I wrote:
The Standish Group original 1994 "Chaos Report" survey of 365 IT managers found 31% of IT projects were "impaired" (canceled before completion or not used) and 53% were "challenged" (late, over budget or failed to deliver the expected benefits). Another 16% "succeeded -- came in on time, on budget, and had what it ought to have," according to Johnson. The trouble is, "people lump the failures with the challenged projects." The latest Chaos Report from 2004 shows an improvement: Only 18% of projects failed, 53% were challenged and 29% succeeded.Other recent studies also offer more encouraging numbers. The 2005 KMPG "Global IT Project Management Survey," involving more than 600 organizations in 22 countries, found only 49 percent of respondents have experienced even a single IT project failure in the past 12 months.
A "challenged" or even a "failed" project is not necessarily a total loss, says Johnson. Challenged projects can still be judged a success by companies at the end of the day if the results are worth the extra expense or time. And, says Johnson, "There needs to be a certain amount of failures. When people say they have 100 percent success rates, either they are lying or they haven't pushed the envelope far enough."
Going by that, I don't think the 17 percent over budget figure -- if we don't count accepted changes in project scope -- is unreasonable, even if it's surprising.
