Tina's team had put a lot of work into their estimate for the Giga-Bill 1.0 project, which they had estimated would probably take 12 months. Her boss, Bill, wasn't happy with the estimate the team came up with. He said it needed to be shorter. Tina found herself sitting across from Bill at the oversight committee meeting.
"The team has estimated it can deliver the product in 6 months," Bill said.
"Err-ahem." Tina cleared her throat. "What Bill means is that we estimated an ideal-world, best case of 6 months, but in order to achieve that best case, every single thing on the project has to go perfectly. And you know software projects-nothing ever goes perfectly. Our most likely estimate is 12 months, with a realistic range of 10 to 15 months." Tina was sweating and wished she had a handkerchief to wipe her forehead.
Catherine from accounting spoke up. "We really were hoping for something shorter. Can't you shorten your estimate?"
"I wish I could," Tina said. "But my team has been over this ground carefully. If I shortened my estimate, it wouldn't be worth the paper it's printed on. Giving you a shorter estimate wouldn't make the project take less time; it would just guarantee that it would be late. There is still a lot of flexibility in this product concept, and as we refine our product concept we can refine it with the schedule in mind." She moved into a discussion of the estimate-convergence curve and felt some relief at being on familiar ground.
"This process of refinement isn't entirely neutral," Tina concluded. "We can continue to work with the product concept and resources to shorten the schedule. There are lots of things we can do." She explained a few of the degrees of freedom.
Committee members asked several questions about specific options Tina had proposed, and they seemed satisfied with her answers.
"I'm going to need to think this over," Catherine said. "Twelve months is a long time, but you've given us a lot of interesting possibilities." Tina invited her to call with questions or to discuss more options.
After the meeting, Bill was still steaming. "Don't mess with me again," he growled at Tina. "Where do you get off changing my estimate in front of the committee?"
"Your 'estimate'?" Tina replied. "You didn't have an estimate. You had a schedule target, and that target happened to be impossible. My team isn't just throwing darts at a board. They did a lot of work to come up with an unusually solid estimate, and 12 months is the best number we have for this project. This organization, including you, has a history of overrunning its schedule and budget targets, and I couldn't let you set yourself up for failure. I was careful not to make you look bad. The committee took the news pretty well, I thought."
"They did take it better than I expected," Bill admitted. "I'll let it slide this time. That was a gutsy move, but don't ever pull a stunt like that again."
"OK," Tina agreed, wondering whether she would ever need to pull a stunt like that again. Her stomach had done flip flops when she corrected Bill during the meeting, but she knew if she didn't do it then, she'd just have to do it 9 months later when it became obvious that they weren't going to meet their schedule. If she waited until later, the next 9 months would be a miserable exercise in code-like-hell development. Bad planning and poor quality would push out their real deadline, and they probably wouldn't even deliver in 12 months. On the whole, she thought she'd made the right call.