Projects that work: being on time, within budget, and to specification can be difficult. As one project manager explained, "you can have it fast, cheap, or good. Pick two.".

• Author: Graham, Andrew
• Position: Business & Finance

EACH YEAR, an estimated 10 trillion dollars is spent on projects around the world, many of which fail. In fact, there hardly is a company that is satisfied with its project performance. By some measures, two-thirds or more new product, technology, and merger projects fail to meet expectations. Why, given the high rate of failure, do companies keep trying? Surely, there must be an easier way to earn a living.

People have been taking on projects for thousands of years. Long before project management software, Earned Value Analysis, and Gantt charts, Egyptians were building pyramids, Romans aqueducts, and Americans and Canadians transcontinental railroads. We keep starting projects for the simple reason that things have to get done. Needs must be met. Greater value always is sought.

Behind every project stands a core objective--"why" it should be done. The ultimate test of success is whether a project meets that objective and produces the intended value.

It is critical to understand that whatever projects an organization decides to undertake reflect prior decisions as to the best way to meet a need. For example, launching a new product is just one way to increase market share; merging two companies is one method of advancing company strategy; redesigning a process is one of many ways to reduce costs or raise quality. In most cases, there are a number of alternative projects that will meet a company's objectives, and the one that management decides to undertake simply is the one that appears to be the best choice at the time.

In the world of projects, success begins with picking the fight ones. Consider the U.S. invasion of Iraq. A plan to overthrow Saddam Hussein carefully was thought out and executed. His regime fell; the U.S. captured Baghdad and American troops now are trying to maintain the peace. If its goal simply had been to oust a dictator, the project would have been a great success. Yet, if the ultimate goal is to achieve greater stability in the region and make the world safer from terrorist attacks, the project, thus far, has been unsuccessful--although it is far from over. More over, if this was the goal, it raises two important questions. Was the implemented project plan really the best alternative for creating the desired result? Should the project plan have been modified as it came up against reality? History will be the final judge. For now, it is left to leaders, generals, pundits, and the public to debate.

Likewise, if an organization undertakes a project to install an enterprise customer relationship management system, the success of that project will not be defined as "installing X number of workstations, networked with Y number of servers, running Z number of configured software modules at some predetermined level of performance." Success will be defined first by whether that project was the right choice to meet the organization's business objectives, and second by the quality of the countless decisions that need to be made throughout the project: how to scope and organize it, whom to assign work to, how risk should be managed, and how to respond to the challenges that arise during implementation.

Choosing the fight projects on which to work requires understanding the organization's needs. Every project should, in some way, address one or more of three strategic needs by adding market share, size, or margin. That's it. Any project, whether it is acquiring a company, launching a product, reducing cost, increasing customer satisfaction, raising employee retention, upping shareholder value, or even...