Networking

Preparation, time buffers help prevent project bottlenecks

Predicting and preventing bottlenecks is like playing chess--you're wise to think two or three steps ahead. Check out some examples of projects whose progress was helped by anticipating and preventing bottlenecks.


Part of managing a project or an organization is positioning yourself to anticipate what might go wrong. Problems like losing key staff members or the failure of a member of a project team to complete key tasks on time can cause the project to take longer, cost more money, or fail to meet the objectives.

It’s your responsibility as the leader to find the bottlenecks, or constraints, that can potentially jeopardize your objectives. I take the position that my staff, or project team, may not recognize potential problems. That doesn’t mean they’re weak or incapable, they just may not have experienced the situation in a context that allows them to recognize the potential for an issue to become a problem. Bottleneck issues can come in many forms, including:
  • Staffing
  • Budget
  • Client dependencies
  • Vendor dependencies
  • Project dependencies
  • Task completions
  • People issues
  • Process dependencies

You can even add weather or natural disaster issues to the list. The point is that you have to be constantly on the lookout for any conceivable bottleneck. Keep your eyes and ears open and look beneath the subtle messages you see and hear as you work through a project or go about your daily business.

Planning for holdups
Civil War generals positioned themselves on high ground so they could see the battle develop and anticipate the opposing side’s moves. By observing and anticipating, the better generals won their battles, often against much larger opposing forces. The same strategy works for consultants: Keep your senses on high alert for issues that can cause you problems or jeopardize your mission. I’ll give you a few examples.

In a recent systems conversion project, we had to get the WAN circuit connections live in seven remote offices before we could conduct user training on the new system. Because I knew from experience the uncertainty in dealing with telecom companies, we built in extra time to ensure we wouldn’t jeopardize training. Rescheduling training for all those employees would have been a real problem for the project.

The challenge with building in extra time is that your staff often feels like they have plenty of time to maneuver. Knowing that, I asked my infrastructure leader to give me a daily update on the connectivity status of each office. If this objective wasn’t moving as I thought it needed to, I emphasized the need to communicate the sense of urgency to the telecom companies.

I always build my plan to have WAN connectivity up at least 30 days in advance of when I really need it. It costs me a bit extra but saves me time and worry in the long run.

A second example is part of a process that I often build in to the PC fulfillment for new employees of a company. Too often, a manager forgets to let me know that a new employee is starting and will then request a PC at the last minute. In several companies, I eliminated my fulfillment bottleneck by maintaining a couple of spare PCs that were already imaged and prepped for deployment.

The new PC order justification worked as it always did. The difference was that I could immediately fulfill an order when the last-minute need arose. Some would say that this only enables poor managers to continue forgetting their responsibilities. My advice is to have the conversation with managers about improving their actions, but in the meantime, get the new employee started on the right foot.

Here’s a third example. During the mid '80s, I managed a small organization supporting 26 hospitals. One of our services was providing dedicated programmers to clients that had a big programming need and could justify the cost. I’m sure my staff thought I was daydreaming at times when they walked by my office and saw me staring at the wall. I was actually anticipating how I would staff the next two or three placements of a dedicated programmer, when I needed to hire, and what my recourse would be if certain events developed. The point is that to eliminate bottlenecks, you have to anticipate them, find them, and have a strategy to address them when and if they pop up.

Eliminate your number-one bottleneck and two take its place
The book The Goal: A Process of Ongoing Improvement has a great section on identifying and eliminating bottlenecks. One of the lessons in the book is that as soon as you eliminate your most pressing bottleneck a new one materializes. In fact, it’s possible to create a bottleneck that wouldn’t have existed had you not eliminated the first bottleneck.

Here’s an example: To prevent a bottleneck of a project task, you reassign a team member to assist. The primary bottleneck gets resolved but the reassignment causes another task to go untended, which ultimately becomes the new bottleneck of the day.

Your goal, of course, is to help the project progress. Knowing that new bottlenecks surface when you eliminate others can help you. For example, I now subconsciously anticipate and seek out what I believe are the next three to four bottlenecks of a project. It’s like playing chess and anticipating your opponent’s next three or four moves. For certain, the most important is the primary bottleneck, but knowing which tasks will become the new bottlenecks after the primary one is eliminated helps you plan and take strategic action to help the project make progress.

Remember the example of getting the WAN circuits live so we could conduct training? Well, another bottleneck followed as soon as that one was eliminated. Getting a training library set up for training with the users’ real data was necessary in our environment. It was great to have connectivity, but without the training data and users configured, we couldn’t conduct training. You have probably already figured out that I emphasized getting the training library in place well ahead of training and in tandem with getting connectivity in place, especially since we could do it irrespective of the WAN connectivity tasks.

Make it a point to identify as many of the tasks or issues surrounding your project that may become bottlenecks as you can. Tasks that have a real opportunity of causing the project to miss its budget or delivery date should be addressed quickly.

Options for eliminating bottlenecks
You’ve seen three very different examples of a bottleneck. All were dealt with differently. In general, I try to eliminate a bottleneck in one of the following ways:
  • Use a buffer to allow for certain tasks to take longer than expected. This action prevents an issue from becoming a real bottleneck.
  • Implement a workaround.
  • Put additional resources on the issue.
  • Place a higher priority on the issue.
  • Escalate to a higher management level for assistance.
  • Follow up, follow up, and then follow up some more.

Again, your best asset is anticipating what can become a bottleneck and working to prevent it from becoming “the bottleneck” ahead of time. Coaching less experienced employees on how to identify bottlenecks, how to prevent them, and how to eliminate them makes for stronger professionals on your team. It also creates a lot of pride when you deliver your projects on time and within budget.

Mike Sisco is the CEO of MDE Enterprises, an Atlanta IT management training and consulting company. For more of Mike’s management insight, take a look at his IT Manager Development Series.

Editor's Picks