Johanna Rothman makes comment on PMHut as to why managers like serial life cycles. But she misses a critical aspect of all business processes at the macro economic level. In the business world macro economics deals with the performance, structure, and behavior of the business and the projects and processes executed by the business.
So the macro economics question that must be continuously asked and answered is...
What is the budget needed for this project? and How are you performing against this budget?
Unless you work for Daddy Warbucks someone will want to know how much a project will cost and when - more or less - it will be done. And then the big question that business funders want to know is
If I give you the authority to spend within the allocated budget, and you can tell me - more or less - when you'll be done spending my money, will the result be what I asked for?
This is why business managers like serial life cycles. It bounds their risk exposure.
The conjecture that projects cannot be estimated is of course pure conjecture. It's done all the time and a wide variety of domains. Software intensive domains.
The absolute fundamental principle of cost and schedule estimating is that NOT SINGLE POINT ESTIMATE is credible. Johanna makes one of those wonderful inverted logic arguments.
If you use a single point estimate tool you'll be in wrong
Yes Johanna is absolutely correct, but also absolutely wrong. NEVER EVER use sign point estimates for cost, schedule, or technical performance. In in fact in the Us Defense business single point estimates are forbidden. See DID 81650 for the guidance on what is allowed.
Take a look at some suggested approaches to Managing In The Presence of uncertainty:
- Risk Management in Five Easy Pieces
- Increasing the Probability of Project Success
- Building a Credible Performance Measurement Baseline