When Bad Things Happen to Good ProjectsBy Baselinemag | Posted 2002-06-17 Email Print
WEBINAR: On-demand webcast
Next-Generation Applications Require the Power and Performance of Next-Generation Workstations REGISTER >
Is your project at risk for disaster? Learn the threats you face.
If all projects were the same, life would be easy. Whoever ran one project could be handed the next, with success virtually ensured. In reality, projects vary widely, as do the skills needed to manage themso argue three professors in "Managing Project Uncertainty," a Sloan Management Review article (Winter 2002) that postulates a continuum of threats, from variation to chaos. Below, we outline the threats, using Baseline case studies to show how managers performed in the real world. (Some did superbly; others flunked.) To get a sense of the threats your project faces, take our quiz.
Almost every project is subject to delays and cost overruns brought about by minor factorsworker illness, supplier delays and the like. Slip happens.
The ability to spot problems early and resolve them quickly. Smart managers sometimes insert cost and schedule buffers to counter the impact of variation.
Project: Deployment of resource software in all divisions of multibillion-dollar company
What happened: In 1994, $5.7B apparel-maker VF Corp. embarked on a five-year plan to install SAP's resource-planning software across the entire company (Baseline, Case 003, October 2001). It didn't work out that way. The rollout to the first division took 15 months longer than expected when promised software modules didn't materialize. More recently, VF abandoned a year-old effort to implement SAP at its second-biggest division, because that group lacked the resources to do the integration work.
Project team grade: B-
The steps VF took to address SAP's limitations (such as doing some custom development) kept the initial deployment from falling even further behind schedule. But that experience should have kept VF from trying to deploy SAP at a division with fewer resources.
Threats are major but easy to identifylike the multiple hurricanes in The Perfect Storm. The only question: Will it happen?
Is it true only the paranoid survive? It is with these projects. Best suited: those with the intuition to anticipate obstacles and the discipline to formulate alternate plans ahead of time.
Project: Wireless rollout What happened: When transportation company J.B. Hunt decided in 1997 that it needed to do a better job of tracking the location of its trucks and trailers, the wireless industry was in its infancy (Baseline, Case 006, November/December 2001). The $2B company paid the price for being an early adopterthe three vendors it has used on this project have either gone out of business or restructured. As of last fall, Hunt had only 5,000 vehicles equipped for wireless, a mere third of its target for the time.
Project team grade: C-
Hunt had no Plan Blike subcontracting the development of its wireless system to a consultant that would have assumed the risk of vendor failures in an emerging market.
Many managers say it's the risks they can't see that worry them the most. Sometimes, a succession of moderate risks, each foreseeable on its own, crash down on you all at once.
When you're dealing with gremlins, it helps to have friendsor at least relationships of trust with peers, managers and business partners. The shared dependencies that result give you resources to react to whatever comes.
Project: Tapping an outside consultant to manage a major systems transition
What happened: Collectibles-maker Department 56 was following a well-established tradition when it hired Arthur Andersen in 1996 to help upgrade its J.D. Edwards software package (Baseline, Case 014, March 2002). But using a consultant doesn't mean you can take a vacation from your project. Not realizing that deploying ERP takes longer than advertised was Department 56's first mistake. Its second was not entering 1999 price and inventory data into its old system, forcing it to cut over on the go-live date of Jan. 1, 1999, ready or not. The $200M company paid for its errorsfor not staying on top of Andersen, in particularwith a screwed-up system that cost it millions in lost sales and bruised its image. It got $11 million out of Andersen in a lawsuit earlier this yeartoo little, too late.
Project team grade: F-
(Summer session for these guys.) It's always a bad idea to put all your eggs in one basketbut if you do, you should take Mark Twain's advice and "watch that basket." Department 56 didn't.
Unforeseen events may invalidate goals, over and over. These projects often have no clear structure or definable endpoint.
Sudden, dramatic changes in the dynamics of a situation call for improvisational skills. You learn rapidly and adaptor you fail.
Project: Recovering data and applications under extreme time pressure
What happened: Despite the deaths of 180 of its technology workers in the Sept. 11 World Trade Center attack, trading firm eSpeed had its service up and running by the time the bond market reopened on Sept. 13 (Baseline, Case 002, October 2001).
Project team grade: A
Yes, the project team had disaster-recovery plans in place (it was prepared, in short, for a foreseen uncertainty). But no one could have predicted the scale of this disaster. The Cantor Fitzgerald subsidiary's creativityrouting calls through its London office when domestic circuits were not available, for instancewas clearly entrepreneurial.