Measuring the Success of Projects, Within Tolerances
All projects should establish a scorecard that describes what it means to be successful. This should include project management metrics covering estimated effort, project duration and cost, as well as client satisfaction with the process. It should also include technical metrics such as defect rates, rework targets, and other important product characteristics. When you are defining your metrics, make sure you build in the idea of target tolerances.
Tolerances are a way to build in "reasonableness" - there is no such thing as perfection in project management! Your organization should establish the tolerance levels that they consider acceptable for project management. For instance, a normal tolerance range for a typical project might be plus or minus 10%. That is, if you delivered the project for no more than 10% over budget, it is still considered a success.
The problem with setting reasonable tolerance targets, however, is that project managers may come to conclude that, in addition to a contingency allowance, they have a sort of "unofficial" allowance of another 10%. Another issue is that any such "forgiveness" should be tied to the risk level of the project in question. It is not reasonable that a project involving entirely new technology should be tied to the same tolerance as a standard run-of-the-mill type project.
Declaring success from a project management perspective is normally what the project team is asked to be accountable for. But from a project portfolio perspective, it is the quality and performance of the product that brings value to the organization. So, the ultimate issue is whether the organization received the value that was promised from the original benefit projections.
However, whether the expected benefits are actually harvested from the product usually depends, not on the project team, but on Operations management. That's why the project team is typically held to the narrower goals of project management. Nevertheless, if the project was a failure from a project perspective, the chances are that it will also prove to be a failure from a corporate perspective as well. (This is not always the case. You may complete a project over budget and schedule, but the organization may still gain value over the long-term lifetime of the product.)
Conversely, there are also many examples of projects that were successfully delivered, yet are not delivering the value promised. If the project team delivered successfully within tolerances, there is usually nothing else that can be done from their perspective. However, in a portfolio, the overall business value derived from its projects should be monitored over time by the portfolio team after each project has been completed.