Comments on _Critical Chain_, ch 8

The book gets back to the issue of researching projects to study the safety of the steps.

He locates it quickly. "Okay. 'Due to budget overruns (sixteen point two percent) and delays in production, the original estimate of three years payback is now modified to five.' Typical. What's your point?"

"The budget overrun, being only sixteen point two percent, cannot possibly change the original estimate for the payback period by more than half a year."

"So?”

"But they had to increase the estimated payback period from three years to five. By the way, the person writing it is a project auditor and he claims that his friends are already pushing to change the official estimate to seven years."

Jim still doesn't get it. It's not like him. Patiently, I continue, "If the budget overrun can't possibly cause such a change in the payback period, it must be caused mainly by the delays in completing the project."

This was foreshadowed earlier in the book (chapter 6). It was said that in production, in contrast to projects, there's a lot of waiting in queues. 

“Yes, but that's because production is different," Fred argues. "In production, most of the time parts spend in the plant they are waiting in queues in front of machines, or waiting for another part in front of assembly. Most of the lead time is not actual production, it's in wait and queue. That's not the case in projects."

Continuing...

“In the same report," I don't give up, "it's indicated that they chose the cheap vendors over the more reliable ones. How much do you think they saved?"

"How do I know? Maybe five percent. Can't be much more."

"You can also see," I continue, "that delays in getting the machines from those vendors was the prime reason for the delay in completing the project.”

ugh. So they accepted a trade off. Get a 5% savings on the machines, while accepting an extra 2 years of added time to the whole project (from 3 years to 5 years). This is dumb as hell because the extra 2 years represents far more money than the 5% savings figure. The 2 years of extra time costs money in the form of 2 years worth of lost profits. Using hypothetical numbers, imagine a $100,000,000 dollar project with an expected rate of return of 15% and the 5% savings on the machines represents $3,000,000. The 2 year overrun constitutes a loss of profit of more than $30,000,000. What a crappy trade off.

This is the mentality that Goldratt talked about in other books: people routinely focus on optimizing parts while sacrificing the whole.

It's weird. A famous saying in America is "time is money", but it seems like these projects don't treat time as money.

Maybe overruns should be linked with extra fees. So like the customer would get extra fees if there's an overrun, helping recover their loss in the case of an overrun. And the company's incentives would better match the reality of the situation.

And the company's sales team could position themselves above the competition by guaranteeing no overruns or else hefty fees would be added.