First, Earned Value formulas do assume a linear relationship, and frankly, what sort of non-linear relationship would you choose in its place? (I can’t think of one…)
Second, the 80 20 rule is indeed an almost universal rule of nature, but I would say that it generally occurs for ‘open-loop’ nature. When you have a project manager in the loop, performing good estimating (that’s a topic on its own…), and taking controlling action, the idea is to stop the ‘80% done’ syndrome.
But back to earned value for a moment.
The ratios of efficiency and effectiveness used to forecast the EV future are based on work performance thus far, and for a given team, that makes reasonable sense to me. EV also uses cumulative formula to use trends based on past team performance as well, so I continue to believe that the PMP Exam takes the best from EV.
But EV is a powerful tool, but the PMP exam expects you to be able to understand and use the basics – quite right too…
I’m currently knee-deep in creating an Agile Primer, and I’m reminded (just like the PRINCE2 Method), that the 80/20 effect is usually caused by NOT planning and measuring Products/deliverables AND Activities together.
Agile techniques include the powerful Timebox approach where the requirements/product/deliverable/features – call them what you will, are prioritized so that the Must Have’s are guaranteed and the Should Have’s and Could Haves are expected. Tight control around these ensure that Timebox delivery times never succumb to the 80 20 rule. For the purpose of this article, assume an Agile Timebox is the same as a traditional project’s Work Package.
My final response to an interesting question is that one of the reasons why 80/20 occurs is because at some point during a task, the team/project manager realize they underestimated the WORK EFFORT, at this gets artificially ‘bolted on’ at the task end – thus ‘proving’ the 80 20 rule.
In a better world using both product and task planning for Agile Time-boxes or traditional project Work Packages, a more accurate work plan can be created (and measured/forecasted using EVA). Linear EV will get the job done.
Thinking about the typical situation when 80?20 raises it’s ugly head, is when the team forecasts that they will be finished by a certain date, and the project manager discovers otherwise about 24 hours before the task was due to be completed.
The drivers for this happening are usually:
1. The estimates only used work effort
2. The planning only used task work
3. The correct skill/knowledge/experience for both estimating and delivering were not present on the task!
4. The sub-task deliverables or products AND their quality/acceptance criteria were not established and included in the plan
5. Risks were not considered and managed properly
6. Agreed reporting/audit/reviews were not included within the task – hence the end task 80/20 crisis
So, a very long answer for a simple question. Don’t try to use non-linear formula to second guess human nature, do proper planning at the right level of detail needed for accurate task control, and use Earned Value Management Analysis to assist such control.
David spent 25 years as a senior project manager for USA multinationals, and has deep experience in project management. He now develops a wide range of project-related downloadable video training products under the Primer and PM Mastery System brand names. In addition, David runs project management training seminars across the world, and is a prolific writer on the many topics of project management.
PMP – the world´s most sought-after Project Management Qualification
Don’t Let ’em Steer Your Ship Skipper
Meeting the PMP Exam prerequisites
How To Meet the PMP Exam prerequisites
Control of Communications
PMBOK Professional Integrity
Project Charter Example