Arguments against "top-level" schedule risk analysis

[Rick Williams]

There are a number of highly-paid consultants in the project management field
who make their living by convincing management that they can give them
valuable assistance in managing their project by performing a so-called
"top-level" schedule risk analysis. I have been doing schedule risk analysis
for over 20 years, and I believe in doing them "the hard way" - I use the
most detailed schedules I can get, and have the people who are doing the work
estimate optimistic and pessimistic durations, then use a Monte Carlo
simulation tool to find all the possible critical paths, and provide
management with detailed information on what areas of the project should
receive the most management attention and resources in order to increase the
probability of success. And I am good at it.

But still, management often falls for the claims of these outside
consultants. They come in and interview a handful of people and try to find
out what people think are the high-risk areas in the project. Then they build
a small schedule network and use some Monte Carlo tool on it, then they write
their report. In every case I have seen, this report simply regurgitates the
information from the interviews - garbage in, garbage out. They could just as
well have skipped the Monte Carlo analysis part - they are using it to make
it look like their answer is "scientifically" derived.

So, how does one argue against this and keep management from wasting money?

 

[Dale Howard [MVP]]

Rick --

I applaud the steps you take to plan and manage the risks associated your
projects. You have done a very good job of "tooting your own horn" in this
newsgroup. I think most of us would enjoy working with a skilled
professional such as yourself.

My only question is whether you have "tooted your own horn" for your bosses.
Why would they consider using a consultant and paying big bucks for someone
to do less than you do in risk planning for your projects? Oops! I forgot
that Dilbert is not made up: it's almost always based on real world
experiences of people like you! Seriously, have you discussed this
issue with your bosses? And if so, what do they say in response?

 

[Jim Aksel]

To echo Dale's advice, you could also take one of these "High Level"
assessments and you should be able to shred it in seconds.

If you are a consultant, use an example in your sales pitch. If you are
working for a company, you need to show your management where the high level
risk analysis falls apart. And, for the record, we see eye to eye on this --
bottoms up is the only way to get a good answer assuming the triple point
estimates are developed properly.
--
If this post was helpful, please consider rating it.

Jim Aksel, MVP

Check out my blog for more information:
http://www.msprojectblog.com

 

[Andy Boruta]

Even with the bottom-up approach, the quality of the assessment lies with the
accuracy of the estimates.

How do you assure that all estimates are calculated using a common set of
assumptions? I am assuming that different people are estimating different
activities for you.