allPM.com :: Troubled Projects: Identification, 5-Step Action Plan or Closeout By Margaret Rubick

Troubled Projects: Identification, 5-Step Action Plan or Closeout By Margaret Rubick

Published on Friday, November 30, 2007 - 02:42 PM

I spent fourteen years at IBM managing projects, some internal, some external, and some multinational. The questions “What constitutes a trouble project?” and “What do we do with a troubled project?” cannot be asked too many times. How do you recognize the warning signs? Sometimes it’s a missed deadline, sometimes it’s a gut feeling of discomfort, and sometimes it’s the intuition that no matter what you do, you will have a dissatisfied client. Learning to identify troubled projects early and coming up with strategies to deal with them can save us from a lot of sleepless nights.

The reasons for troubled projects are legion. A few possibilities, all of which I have encountered:

the sales team sold a project optimistically, wanting to close a deal. Now the delivery team has to live with it and can’t see a way to deliver.
Scope creep: you don’t want to “nickel and dime” a client who has agreed to pay millions of dollars for requirements or for an installation, so you say “Yes” a few times too many.
Estimates are notoriously low—for all aspects of design and delivery. We all like to think we can deliver more than is humanly possible, and we don’t usually like to acknowledge that we are susceptible to the flu or have administrative work that is separate and in addition to our committed project.
We have a project team that is agreeable but not dedicated. Members cannot allocate all their hours to the effort we’re responsible for. We ask, cajole, beg for attention of a team that is stretched too thin.

Five steps you can take:

Determine what’s broken—not whom to blame. I always asked my teams to raise issues early and promised to not “shoot the messenger.” If we identify problems early, we can fix them. Whether you choose to use the fishbone, root cause analysis or a brainstorming session, keep asking until the problem is distilled to no more than three possibilities.
Identify 2-5 action items that could alleviate the problem. This might mean more resources, more time or a change in specifications. It might mean, in the case of an implementation, that you implement a smaller part of the system, or that you phase installation.
Meet with your client. Bring the problem statement and the recommended action items with you. Ask for their vote on your recommendations. Find out what their preference is—is it saving time or money or avoiding risk?
Regroup with your team.
Relaunch.

There is a strong inclination when a project is underway to just keep pushing, work longer hours, and try to “make it happen.” Again, if you’re talking about an event that has a fixed date, with millions watching, that may be the only option. You bring on more team members, send more people out to the field. You manage in crisis mode, and this can be done for a number of months if necessary. It is not a good long-term strategy, however. Team members burn out, mistakes multiply. If at all possible, it is good to freeze for a day or two or three, make the necessary changes, and then relaunch.

I managed a project once for a client that I won’t name here. It was an installation project at multiple sites over a broad geographic area within a short time frame. Installations were running late, team members were stressed. The biggest problems? 1) There was incomplete information on the sites, 2) there was a problem tracking equipment. I was brought onto the project when it was already underway and already troubled. What to do? Team members needed time and better information. Too much time was being spent fixing problems and not enough was left to forecast and plan. We froze the project. It was a calculated risk and at first it was not a popular decision, but it was the right decision to make. The team was able to put together a comprehensive system to track inventory and shipments. The time taken to regroup was regained in efficiency moving forward. The customer was happy with the results.

Projects: save or close out

And for the projects that cannot be saved? There are projects that cannot be brought in both on time and under budget. These are the ones that were sold optimistically, or, in some cases, as “loss leaders,” where the goal is to gain market share or perhaps, test a new process or technology. The most important factor here is clarity. If it costs $2 for every dollar of revenue you gain, then it is a business decision to continue with the effort. If it is an internal project, and there is cost accounting but no profit and loss being tracked, then it is an investment project. There is nothing inherently wrong with an investment project. The trouble comes with expectations—either expectations that a project will make money or show a measurable return on investment (ROI). The most important step out of the five listed above is “Meet with your client.” Whether it be an external client or a shareholder, the investor has the right to make an informed decision, and it is the project reporting that guides that decision. The project team’s role is then to execute or close out the project. One of the biggest challenges then is maintaining morale. It is possible to do excellent work on a project that cannot succeed, and team recognition plays a big part.

In summary, there is no “magic wand” for troubled projects, but there are definite steps that can and should be taken when trouble is brewing. Ask why, identify action steps, communicate, reorient, and relaunch or closeout. Whether running a project, driving a car or sending a rocketship to the moon, we must always check coordinates periodically. Having fun along the way never hurt anyone, either.

© 2007 allPM.com

Margaret Rubick has managed projects, troubled and otherwise, for more than 20 years. Her first project was at Safeway Stores in the mid-80s. A successful network implementation led to work at Pacific Bell in California and then a 14-year position at IBM, beginning as an advisory Project Manager and ending as a Project Executive. Ms. Rubick has been freelancing since 2003, working with a variety of clients in New York. In addition to project management, Ms. Rubick performs business assessments, facilitates meetings, and works one-on-one and with groups to problem-solve. She administers personality assessments and focuses on creating positive outcomes.

How well does management support newly assigned project managers?