Project Management: Have We Started Yet?

Measuring project success or failure from initiation through achieving the business value that was agreed upon may not give us the information we need to improve our delivery. Initiating and planning a project correctly can end in cancellation or continuation into the build phase of the effort. A measure of success for initiating and planning a project correctly could be cancellation. The more precise the metric the better the information we will have for improving our ability to delivery projects and achieving the business value necessary.

One of my favorite books on motorcycling starts out by discussing motorcycle accidents. The reason for starting with the accident seems pretty obvious. If we know why accidents happen then we can develop skills for avoiding those types of situations. They gathered statistics on single and multiple vehicle accidents and categorized the primary cause of the accident. The causes were articulated in simple terms such as people pulling out in front of the motorcyclist, turning left in front of the motorcyclist, gravel in the road and such. While the details of the study are fascinating to me as a motorcyclist I think the value to me as a project manager is not in the outcome of the study but in the study itself. Most important to me is the boundary used for collecting the data.

Most motorcycle rides start before the motorcyclist actually gets on the bike. Choices are made long before pulling out into traffic. If it is a short ride fewer plans are made but the number of choices we’ve made along the way are about the same. The typical decisions are going to be where am I going, what route am I going to take, do I need to have storage available on the bike and what am I going to wear. In order to make those decisions I have to ask some questions (e.g. Do I have enough gasoline? Am I picking anything up along the way? What is the weather like? Is anyone coming with me?). The point I’m making is that the study captured data based on failure during the execution of the ride, not failure during planning or any other phase of the ride.

How is project failure measured? In the Standish Group’s Chaos report of 1995 (used here because of the popularity of use for project failure statistics) project failure was defined as any project that “is cancelled before completion or never implemented”. Should project failure be based on a similar measure as the motorcycle accident study (The Hurt Report)? Is a project a failure if it does not deliver the agreed upon service or product? How should we define that delivery, simply implementing the product or service based on the standard project delivery metrics (time, quality, and scope) or achieving the business value for which it was undertaken? And when do we start measuring? Is there value in measuring the projects that start and never get to the “build” phase which correlates to the actual act of riding the motorcycle? Using the Standish Group definition, a project could be cancelled before the project gets to the build phase and be considered a project failure. Why does it matter how we define the start and end for project failure?

The metrics we’ve been using have helped define the project management industry as a whole. The methodologies, tools, techniques, certifications and training classes have been in existence to try to solve the problems identified in studies such as the Chaos Report. While project delivery has improved I believe we can improve faster in efficiency and effectiveness if we can agree on the definition of project failure and when a project starts and stops being measured to determine failure (the ride).

I don’t want anyone to get the wrong idea – I believe the project studies have been valuable and have helped the project management industry – project performance has improved. I also believe that we must put in metrics that measure a projects success or failure based on delivering the business value, not in measuring time, cost and scope. While these may be delivered successfully, if the effort does not deliver the value it was designed to deliver then the project is not successful. That would also mean that we would need to be clear on when we call a project a measurable project. We don’t measure motorcycle accidents as any time a person thinks they are going to take a ride and decide against it. Why would we measure project failure based on thinking a project is a good idea and deciding differently? Let’s start measuring projects when they get “on the road”? Thinking about the motorcycle it would mean that someone had an idea, did some analysis, planned it out, got everything ready and then started the motorcycle. I think project failure and success should be measured based on “starting the motorcycle”. In a software project it would mean that the initiation, analysis and a majority of the design would be done prior to measuring the success or failure of the project. Is that too late in the process?

I have some concerns about measuring a project for success or failure following the completion of the high-level design. My main concerns are how much time, energy and money are spent doing those activities. How do we ensure that the effort is not excessive or wasteful? In other words, how do we end the wrong projects quickly and keep the right projects, the ones that will succeed, moving along quickly? I don’t think we can gain the insight into those questions measuring a project from inception through ROI.

I think measuring project success/failure based on “the ride” or “the build” portion of the project through ROI is critical. I also believe that the inception of a project through to the start of the build (the start of project failure/success) is part of business operations and therefore measured differently than the remainder of the project. Have we started yet? I would argue, not until you get on the motorcycle.

Ride On, Manage On