Customers and senior managers make impossible demands. Team members produce aggressive estimates. People make optimistic promises they hope to keep. Project managers make plans, and the project gets underway. Then reality steps in. Projects never go exactly as intended and the original plans never last for long.

I’m not all that crazy about reality but it’s all I’ve got, so I have to deal with it. And the reality is that project managers must temper the expectations of other stakeholders, take the estimates they hear with a grain of salt, and be cautious not to overcommit and then underdeliver. Part III of this book provides some recommendations about how to deal with the fact that the future is not entirely predictable and dreams rarely come true.

Too many software projects are launched with the hope that productivity magic will happen, despite previous experience to the contrary. Sometimes a team does get lucky or pulls off a miracle. More often, though, miracles don’t happen, leaving managers, customers, and developers frustrated and disappointed. Successful projects are based on realistic commitments, not fantasies and empty promises. Chapter 9 addresses the importance of "Negotiating Achievable Commitments." It describes how to improve your ability to make sensible commitments both within the project team and to external stakeholders. Building a commitment culture in your team is an important part of project initiation.

People usually attempt to estimate their work in good faith and they make promises with the best of intentions. Sometimes, though, surprising things happen. Requirements grow, risks materialize, team members depart, new markets are identified, and estimates turn out to deviate far from reality. Chapter 10, helps you deal with the uncertainties and risks that lead to overruns by describing how to incorporate contingency buffers (a.k.a. management reserve) into project schedules. The guidance in this chapter can help you avoid your schedule being trashed the first time anything unexpected takes place.

If you were to ask five team members to produce an estimate for a specific body of work you might get five wildly different responses. Which do you believe? The estimators are all making different assumptions and performing different analyses regarding exactly what work is required to complete that project. A better estimation strategy is to put their brains together. The Wideband Delphi method for team estimation is the topic of Chapter 11. Wideband Delphi is a way for a group of estimators to combine task estimates from several participants and lists of their estimation assumptions. The stages in the Delphi process are discussed in detail in this chapter.