The "Death March" Project
Are you managing the project from hell?
12/1/98
by Sara Fleming

It's a scenario familiar to many people: a gray-suited VP or account rep shows up in your office one day with the project from hell. Develop a Web product with a list of features and functions as long as your arm, and do it in half the time, with a too-small staff and an insufficient budget. Welcome to the death march project.

Software consultant Edward Yourdon coined the phrase "death march project" to describe a development project whose "likelihood of failure is >50 percent,"^[1] based on risk-assessment methods standard in project management. It's true that assessing project risk is a lot like herding cats. Problems can develop on any project at any time. Unanticipated market events can put the project at risk. Or your key programmer goes out on maternity leave. The client declares bankruptcy. Anything can happen — but on a well-planned project, the risks are anticipated and minimized. Death march projects, on the other hand, have trouble built in from the very beginning. The project is built on limitations; problems are part of its foundation. On the death march project, survival may qualify as success.

So why get involved? In some cases you may be fortunate enough to see the project coming from a mile away and, better still, you may be in a position to decline to take part. On the other hand, the project may be critical to the success of the company — or your employment there. The death march project may be motivated by seemingly sound business logic: get a product to market before the competition, acquire a major new client, break new technological ground. Or maybe the CEO had an epiphany while stuck in traffic in his new Volvo S-80. Either way, the project's a done deal, and you're stuck on it.

Now what?

What's to Be Done?

The project manager plays a pivotal role on the death march project — and is in the most difficult position of all the team members. He or she can do many things to bring the project closer to success and keep the team on this side of sanity. A project manager's efforts to balance the success of the project and the well-being of the project team are made doubly difficult by the risks and stresses of death march projects. Complicating this is the fact that the project manager will ultimately be held responsible for the project's success or failure. No pressure!

So what can a project manager do to help his or her team survive a death march project, or perhaps even achieve a measure of success?

• Question authority. This is often no easy task, but at first sign of a death march project, the project manager should immediately pay a visit to upper management and explain the constraints, if only to get objections on record. If there's room to negotiate for more resources, a bigger budget, a more realistic schedule, or anything else that will help the project succeed, now is the time to go in with your strongest arguments and get as many concessions as you can. If the project is important enough to management to put it on a death march schedule, it's important enough to put the appropriate resources behind it.

• Then question everything else. Before the project begins, negotiate special treatment for the project team. Make it clear that in order for the project to succeed, the usual corporate rules may not apply to the team. Get the best developers to join the project, and ask for perks for the team: offices with doors, quiet areas, special equipment or software, a testing lab, more comfortable desk chairs, freedom from administrative duties, a free and constant supply of soda, anything small or not-so-small that increases speed, productivity, or enthusiasm.

• Keep it simple. If you're not working from a project spec of some kind, build one. Pare the project down to essential functionality, and then review the spec with the team. Every team member has to feel that they're building a product with enough quality to make it worth bothering with. Strike a balance between a quality product and a streamlined set of features and functions that will help the team get the project finished. Keep a running list of extras. If there's time to build them in, you lose nothing by doing so — after the essential pieces are built.

• Remove roadblocks. As a project manager on a death march project, you should do everything in your power to remove obstacles to the success of the project team. Be utterly ruthless. Start by asking team members to identify the things that slow them down or make their jobs more difficult. Then swiftly set out to remove those roadblocks. They might be minor administrative duties, smaller projects that divide the developer's time, constant interruptions from other staff or management, equipment or software problems, noisy cubicle "neighborhoods," or (the usual culprit) meetings.

• Be a buffer. Protect the team members from interruptions that deflect them from development. Encourage team members to "go underground" — to let the phone ring, and read email only a couple of times a day instead of every few minutes. Put a moratorium on all but the most essential meetings, especially if the meetings aren't directly related to the project. Most important, buffer the team members from upper managers, especially upper micromanagers.

• Contribute an equal share. If the project's team members work late every night, or on weekends, don't walk out the door at 5:00. Nothing will demoralize the team faster, or undermine your team leadership more, than if the team sees you putting in less effort than they are. Pitch in with HTML, do QA & testing, help wherever you can. On the flip side, help team members to pace themselves. Get everyone out the door by 5:00 p.m. on Fridays, as a rule.

• Measure progress. With a death march project, it's particularly important to establish frequent checkpoints in order to meet the deadline. Keep the team in the loop about progress, and make sure you're getting updates from team members about how things are going. Act immediately to resolve any problems — technical, interpersonal, or otherwise.

• Reward achievement. How do you reward achievement on a project that by definition has a 50% chance of failure? Start by rewarding effort as well as success. Save the biggest rewards (bonuses, extra vacation time, promotions, or even new office space) for the successful completion of the project. Then give smaller rewards during the course of the project: buy lunches for the team, get them those little perks like free sodas or new hardware, give them a day off when they need it. Be as creative and as generous as your company's environment will allow.

• Don't let it happen again. Let's say the best outcome happens, and you survive the death march project. Guess what you just proved to those card-carrying MBAs upstairs? Yup, you proved that death march projects can be done. This may mean more such projects to come. Do your best to prevent this by making it clear that the team's performance during the project was above and beyond the call of duty, not standard operating procedure. If the death march still doesn't stop, the last step may be to dust off your resume and find a saner working environment.

Resources

When it comes to project management, the Web industry can learn a lot from software development. The following are several good resources from the software development literature that address issues involved in managing Web development projects.

Edward Yourdon, Death March: The Complete Software Developer's Guide to Surviving "Mission Impossible" Projects. (Prentice Hall, 1997)

Steve Maguire, Debugging the Development Process: Practical Strategies for Staying Focused, Hitting Ship Dates, and Building Solid Teams. (Microsoft Press, 1994)

Steve McConnell, Software Project Survival Guide. (Microsoft Press, 1997)

Steve McConnell, Rapid Development: Taming Wild Software Schedules. (Microsoft Press, 1996)

Tom DeMarco and Timothy Lister, Peopleware: Productive Projects and Teams. (Dorset House, 1987)

For a related article, try Managing Web Projects: Understanding project management for the Web.

Tell us about your nightmare project (or one you pulled out of the fire) in the discussion forums.

Sara Fleming is an Internet Project Manager based in the Boston area.

^[1] Edward Yourdon, Death March: The Complete Software Developer's Guide to Surviving 'Mission Impossible' Projects, Prentice Hall, 1997, p. 4. See the PMI's Guide to the Project Management Body of Knowledge for more on risk assessment. Available on the PMI Web site.

to industry guide

© 1998-1999 Anchor Productions, Inc. All rights reserved.