Try This Before Killing a Problem IT Project

by Bruce Skaistis


You know the story. A major enterprise IT initiative kicks off with lots of fanfare and the promise of major benefits. A few months later, the fanfare has been replaced with frustration because the project is behind schedule and over budget.

This isn’t a new problem. Even though a lot of attention has been devoted to improving the performance of major IT initiatives, recent studies have found a high percentage of projects still end up costing a lot more than estimated, taking a lot longer than expected, or never delivering the targeted benefits.

There have been countless stories about companies pouring more money and time into problem projects long after it was clear the projects should have been killed. With all the coverage of problem projects running amuck, I sense companies have gone to the other extreme and are pulling the trigger very quickly on major IT projects that run into problems.

Out of control IT projects can cost companies big bucks and create all sorts of problems, but scrapping problem projects prematurely can also be very costly and create other problems. I think an effort should be made to get a problem IT project back on track before making the decision to kill the project.


How do you get a problem IT project back on track? Here’s the approach I use when I help companies salvage problem projects.

Start With an Open Mind
Depending on how close you have been to the project, you may already have ideas about what has gone wrong on the project and what should be done with the project – but that’s not where you need to start.

If you are given the challenge of trying to get a problem project back on track, you need to forget everything you think you know about the project. If you go into the effort with an open mind, you are probably going to find a lot of the perceptions you had about the project were wrong.

Identify Specific Problems
The next step is to understand why the project ran into problems. This is not an effort to place blame or find scapegoats. You need to clearly understand the problems with the project so you can determine if the problems can be addressed effectively to get the project back on track.

The assessment is actually fun because you get to play detective. On TV, interrogation scenes frequently involve a “good cop” and a “bad cop.” I’ve found the “good cop” approach works best in determining what went wrong on a project. Conducting the assessment in a positive and non- intimidating manner will encourage people to be open and honest with you in discussing the project and project problems.

One more thing, during this assessment, you have to listen carefully for things that aren’t being said. I frequently find out more about problems on a project when I read between the lines or follow up when I sense someone is holding back. It’s amazing how much you can find out about a problem project with the right approach and by asking the right questions.

Develop Realistic Estimate for Completing Project
Now that you understand what happened to get the project in trouble, you need to figure out what it will take to successfully complete the project. That means how much it will cost and how long it will take to fix the problems and complete the remaining project activities.

Since the life or death of the project will be based on this estimate, this estimate has to be realistic. Your name is going to be on this estimate, so you might even want to make this estimate a little high in terms of cost and time to fix and complete the project. You don’t want to find yourself in the position of having to explain cost and schedule overruns if the decision is made to continue the project.

Revalidate Benefits
One last thing before making a final decision on the project. You need to revalidate the benefits associated with the project.

This is a very important step on a problem project. Time has passed since the benefits were originally identified and the project sponsors have had an opportunity to rethink the benefits in light of the problems the project has encountered. From my experience with this type of revalidation process, the benefits associated with the project are either quite a bit higher or lower than the original benefit estimates. The problems on the project may have convinced the project sponsors that the benefits are even greater than they originally estimated – or the problems may have made them realize the project wasn’t going to produce as much benefit.

If the benefits associated with the project have increased, there is obviously going to be more justification for continuing the project. If the benefits have decreased, killing the project may be the only realistic option.

Determine if Project Should be Completed or Killed
Now it is decision time. Should the project be completed or killed?

Based on what it will take to complete the project and the potential benefits the project offers, this decision should be fairly straightforward. If the decision is close, I typically push for completing the project because turning around and successfully completing a problem project can produce a lot of positives – including better performance on major enterprise IT initiatives in the future.

Establish Meaningful Incentives
If the decision is made to complete the project, you need to do everything you can to ensure the project is completed successfully.

The first thing you need to do is establish meaningful incentives for everyone associated with the project. That includes project leaders, project team members, project sponsors, and owners/users of the new system. The promise of incentives for successfully completing the project can overcome a lot of hurdles and resistance frequently encountered on major IT projects.

By the way, "meaningful" in this context doesn’t necessarily mean money. It could mean extra vacation time, an exciting trip, or other gifts. Whatever the incentive, it needs to create a real commitment to the success of the project.

Communicate, Communicate, Communicate
There is a good chance that one of the reasons why the project ran into trouble in the first place is because of communication breakdowns. So the second thing you need to do to ensure the successful completion of the project is to focus on the importance of communication.

Create a well defined structure for communicating within the project team and with project sponsors, system owners and users, and executive management. There is no such thing as over-communicating on a major enterprise IT initiative.

Update Project Management Processes
The exercise of identifying what went wrong on a problem project can be eye-opening. Take what you learned from dissecting the project and update your project management processes to ensure future projects don’t run into the same problems.

The only thing worse than having to turn around or kill a problem project is finding the same problems on another project in the future.


This approach for getting problem projects back on track is not magic. Some problem projects should be killed--but some problem projects can be turned around and successfully completed.

Before making a decision to flush a problem project and the resources that have already been committed to the project down the drain, it’s worth taking a shot at salvaging the project. Successfully completing a problem project can save a company a lot of money – and significantly improve performance on other major enterprise IT projects.

One more thing. Having meaningful incentive and effective communication structures in place on new projects will go a long way in protecting against problem projects in the future.

About the Author
Bruce Skaistis is the founder of eGlobal CIO. He began his career as a consultant with Arthur Andersen and was CIO of a large bank group before forming his own management services firm. He has extensive enterprise IT management, process optimization, and action facilitation experience.

Copyright ©2006-2007 Global CIO, Inc. All rights reserved. Used by permission.

Back to