Project Stage Gating

One of the things I’ve seen over the years is that the most successful project managers I’ve met have based part of their success on making sure that the projects they start have a good chance of finishing.

This is perhaps particularly true when we look at high-tech or IT projects. It amazes me when I arrive into an organization where there are no criteria for choosing to start a project. I courses I’ve been teaching lately, I was asked to pick a couple of key project management concepts that I thought were the most significant in improving an organization’s project management efficiency. I chose two of my favorites: First, how not to start a project at all and second, how to stop a project that shouldn’t continue. Both of these subjects can be handled through the implementation of a project stage-gating process.

The stage-gating process essentially sets a series of criteria for a project to pass through in a number of phases. There are a number of methods promoted in the public domain but the basic idea is pretty common. The first phase is at the earliest point on which resources would have to be expended. Should you even expend resources to scope out the project and make a schedule? If so, the next phase would be a project initiation of some kind; creating a schedule, estimate, risk assessment and baseline. If the project moved forward, there would be some regular review process and finally, some decision point about putting it into production or releasing it to marketing.

For each phase there are deliverables which must be created and tasks to accomplish. This is the ‘stage’ aspect of the process. Then, before the project can move forward to the next ‘stage’, it must pass through a formal review; the ‘gate’.

This method has been adopted by a number of organizations but it is particularly popular in the IT sector where I think it has made a tremendous impact. You may remember reports like the Standish Chaos report or other statistics from the last few years which quoted incredible percentages of failed projects in the IT sector. Some IT organizations have taken such analyses to heart and have instituted very strong stage-gating processes to combat these effects.

Recently I was working with a Canadian bank who has implemented a strong gating process for their New Product Development (NPD) IT projects. The first phase is the idea phase. Someone has an idea for a project. Perhaps it comes from marketing or perhaps from a developer or maybe the idea starts from difficulty with another project. The first gate is fairly easy to pass. It is only for a tiny amount of resources to investigate the idea and to create a rough plan and estimate. The entire expenditure of resources for the first phase might only be a few hours work. All this work is spent preparing the ‘Gate 1’ deliverables. It’s usually all done by one person, and often it’s the person who had the idea in the first place. The deliverables are presented to a formal Gate Committee along with numerous other Gate 1 ideas. Now the Gate Committee reviews the materials and makes a decision on whether to let this idea become a full-blown project. If it does, a more significant amount of resources are allocated to conduct a proof-of-concept phase. Resources would include people, of course, but might also include real dollars for outside assistance, hardware, software or other resources that might be required to prove that the project should move forward. At the end of this phase, the project again passes before an evaluation committee before the decision to commit significant resources to develop the project to its completion.

For some organizations, the stage-gating process ends there and for many, major benefits will have already been realized. Implementing these two stage gates alone will serve to filter out projects which have a low chance of success and that’s enough to increase overall project efficiency by an enormous measure.

To complete the process, however, stage-gates should occur during project development. Either on a regular project-review basis or at specific milestones, the Gate Committee should review the project status. In particular, one of the aspects of a project that is often overlooked is the business value of the project. If market conditions change, if a competitive product is released, if the organization’s economic situation alters, the appropriateness of this project may change dramatically. In many organizations, these aspects of the project are only considered at the very beginning. Implementing this process into the execution phase of the project can save putting good money after bad to complete a project which has a diminished chance of being commercially viable due to change .

Finally, once the project has been developed, there should be a final gate to pass, the decision to place the product into production or to deliver it to marketing or not.

In the bank we have been working with, there are gates at virtually every one of these phases and the success rate at in this IT department is phenomenal. Resources are constantly being re-focused into the most productive projects possible. The on-time delivery rate of projects in this group where budgets are fixed a year in advance is close to 100% – a remarkable achievement in a highly dynamic project environment.

Whether you are in the IT, R&D, High-tech, or heavy industry sectors, project stage-gating can be of benefit to you. Implementing this kind of process does not require complex software or funding.

If you are considering implementing a stage-gating process, here are some guidelines to best practices that you may find useful:

First, make clear definitions of the phases. Make sure everyone knows how to initiate a phase and how to delineate how to recognize the next phase to ensure that the team actually stops at the gate without rushing into the next phase and expending resources which have not been approved.

Next, have a documented set of deliverables with examples for each gate review process. The driving force of each phase will, in part, become the completion of these deliverables. Given that they’ll have such a significant effect, it’s worthwhile to take some time to ensure the list will generate the kinds of review you want.

Gates should be staff by a committee rather than an individual. Make the group an odd number (to resolve ties) and make sure it’s cross-functional to allow for individual interests. The committee cannot play favorites. To keep them on the straight and narrow, have an evaluation or scoring guideline match up directly to the gate review materials.

One of the responsibilities of the committee is not only to review a project on its own merits but also to decide between projects. This can be done by creating a project selection grid with a scoring system based on weighting between criteria. Some of the criteria between which to choose can include: return on investment, risk, alignment with strategic objectives, alignment with existing skills, balance with other projects and, cash requirement among others. There are numerous examples of such criteria on the Internet.

Finally, settings ‘out of bounds’ criteria can allow you to use the stage-gating process during the entire project cycle. Examples might include being late or over budget by a certain percentage or a change in the project team. Whenever the project falls outside one of these variance thresholds or, perhaps at the end of each project phase, the Gate Committee reviews the project to determine if ongoing expenditure of resources is still expected to deliver the expected return on investment.

Whatever methods you use for implementing a stage-gating process, make sure it’s well documented and that the rules are very public. Nothing can be more damaging to this process than applying it in an uneven or unfair manner so publishing the rules and then ensuring everyone plays by them is the key to a successful deployment.

Originally published April 2004 in PMTimes

Leave a Reply