Choose Your (Project Management) Metrics Appropriately

A recent article over on highlighted recent findings from Forrester Research which found that Common Project Management Metrics Doom IT Departments to Failure. According to Forrester, the idea that a project must be on time, on budget, and deliver the initial requirements is problematic and sets up an IT project for failure.

According to Forrester, the problem with these metrics is that they perpetuate the idea that a project is only successful when it is completed according to the initial schedule, budget and requirements — and therefore, that anything less is a failure. Project requirements change for a variety of reasons, and schedules and budgets change during the lifetime of the project based on better information as to effort, complexity and interdependencies, and, thus, the initial plan should not be adhered to if additional information leads to a better understanding of how the project different from initial estimates.

The reality is that, as the project requirements are explored, complexity will increase or decrease, and, more importantly, requirements will change. Some features and functions will be determined to be unimportant while new features and functions will be discovered that have more value than the initial requirements. Thus, even if these cost more, if the ROI is greater than the ROI predicted by the original project plan, the plan should change. This holds true regardless of the application area — operations, CRM, or supply chain. When it comes to IT, the reality is that you never know exactly how long it is going to take, how much it is going to cost, or just how much benefit you’re going to find until it’s done. You can estimate, and if you have an expert help you, the estimate will usually be close, but it won’t be perfect — and this is why may projects fail, because those who don’t understand IT expect that perfect plans exist. There is no perfection in IT project planning. Accepting this is the key to success.

The metrics that should be used are the ones that identify fundamental issues that cause projects to fail: like lack of governance, unrealistic plans, and limited understanding on the part of management. Better metrics are how many milestones are hit (as long as the project plan is updated at each milestone based upon knowledge gained and lessons learned), what percentage of people are using the system at the end of each phase of the rollout (w.r.t. what percentage of people should be using the system), and how many executives are using the system (even if only for reporting purposes).

Furthermore, project management personnel must play a more active role in managing project sponsors’ and business stakeholders’ perceptions of success and failure. Forrester recommends that project management personnel take the following four measures:

  1. Keep Project Steering Committees on Task
    Insure that the steering committee makes decisions in a timely matter and addresses problems and issues as soon as they arrive.
  2. Improve Communication with Project Sponsors
    Keep the sponsors up to date with changes in requirements and the impacts these changes have on the budget and timeline to insure that they see progress and success and not failure.
  3. Improve the Reliability of Project Plans
    Establish best practices for developing plans with significant unknowns. This can help with setting sponsor expectations for reasonable project performance.
  4. Better Communicate Estimates of Cost, Schedule, and Resources
    … and how they are based on current business conditions, current requirements, and current assumptions — and that they could change as the project progresses and understanding is improved.