The project management profession would appear to be in a confused state!
PMI’s 2008 Pulse of the Profession survey shows improved performance from 2006 with over 55% of project completing on time and over 58% on budget. The survey also found a strong correlation between the project management maturity of the organisation and improved project outcomes.
Meanwhile, the new Standish Group report (April 23, 2009) shows a marked decrease in project success rates,
- 32% Successful (On Time, On Budget, Fully Functional) – worst in 5 years
- 44% Challenged (Late, Over Budget, And/Or Less than Promised Functionality)
- 24% Failed (Canceled or never used) – worst in 10 years
These numbers represent a downturn in the success rates from previous studies, as well as a significant increase in the number of failures.
Around the same time as PMI, Human Systems International Ltd and the Association for Project Management (APM – UK) conducted a survey. The results of this survey reported in the May 2009 edition of Project Manager Today which showed that whilst value realisation is a long way from satisfactory it is not as bad as Standish would suggest. The survey showed 48% of organisations do not measure benefits realisation and of the remaining around half achieve more than 80% of the expected benefit and 22% less than 60%.
It’s hard to know what to make of the conflicting data – superficially, it would appear that organisations that employ professional project management staff (APM and PMI members) do better then organisations overall. But even then, the results are not that good.
An alternative view may be the definition of a project with the APM & PMI membership being more focused than the Standish survey. For more on this see: What is a project?
The last alternative is IT projects (surveyed by Standish) are worse on average than projects in general.
Confused???? I certainly am. The real key seems to lie in the area of project management maturity. Maybe OPM3’s time has come at last?? (PMI’s Organizational Project Management Maturity Model).
Pat Weaver
Lynda Bourne
Aavssitedev's Blog
As a “quant” I have some questions:
1. Did the population of sampled projects change its mix?
2. What does challenaged mean really. Our defense and space programs are ALWAYS challenged, it’s just part of the domain. I’d say the same for enterprise ERP.
3. How many of the failed (canceled) were canceled for the right reason?
4. What if full functionality cost 5% more? 10% more? 50% more. 5%,10%, 50% of what baseline anount. $100K, $1m, $100M?
As a quant the Standaish report stinks. Great marketing for Standish training. Bad statistics.
My feeling all 3 surveys are non-representative. APM & PMI focus on select groups of professional PMs, Standish on IT projects.
Pat,
That’s the core statisrical issue. In the 1952 classic How to Lie With Statistics, you just adjust your sample population to produce the statistics you want.
The representation statistics – how representive are the samples of the actual population are never discussed. Without this infomration we can never know what the expected failure rate should be, or how representative the satted failure rate is of the underlying failure rate.
Great work if you can get it.
To get into the crux of the issue (i.e) how representative the statistics is, i think we need to take a look at the soft indicators of the project that cannot be measured directly.
Visit http://technologyandleadership.com/project-estimation/
to understand what are soft indicators and hard indicators in a project