Project Management Central

Please login or join to subscribe to this thread

Topics: PMO, Portfolio Management, Strategy
Stage Gates Cone of Uncertainty
Network:28



I am in the beginning stages of setting up a small PMO in a hospital and medical facility. I am educating the administration on all things project management. One of the items I need advice on is typical cone of uncertainty for stage gate process.
Prior to establishing project management processes the administration wanted to know all the details 100% up front. While we know that is not ideal and could be costly, I need to educate them on the stage gates and explain how the numbers/costs they see in initiation can change. Does anyone have a typical formula they use?
Sort By:
Network:1856



I used it from years. While Barry Bohem´s has created it for software products it works for other products including the PMI has taken it including into the PMBOK. What you can see into the Cone are not stage gates they are phases in your project. The explanation is: estimations depends on available information and time to estimate and because of that estimations always have an inherent error into them. So, when you estimate, the best practice is you publish a range instead an only estimate. Supposse you are into feasibilty. Because the amount of information you have is you get an estimation of 100 you have to multiply it by 0.25 and 4. Then people that are hearing you need to understand that the project will cost between 25 and 400. The cone has the same for time with quit diferent multipliers.
...
1 reply by January Benjamin
May 29, 2019 1:20 PM
January Benjamin
...
Thank you. This is a great reminder of the ranges and helpful as I set the tone.
Network:28



May 29, 2019 1:14 PM
Replying to Sergio Luis Conte
...
I used it from years. While Barry Bohem´s has created it for software products it works for other products including the PMI has taken it including into the PMBOK. What you can see into the Cone are not stage gates they are phases in your project. The explanation is: estimations depends on available information and time to estimate and because of that estimations always have an inherent error into them. So, when you estimate, the best practice is you publish a range instead an only estimate. Supposse you are into feasibilty. Because the amount of information you have is you get an estimation of 100 you have to multiply it by 0.25 and 4. Then people that are hearing you need to understand that the project will cost between 25 and 400. The cone has the same for time with quit diferent multipliers.
Thank you. This is a great reminder of the ranges and helpful as I set the tone.
Network:263



One place I worked used very strict stage gates, but we didn't have a formula. We had a general R&D process and defined what we needed to know at each gate. The executives were involved in the definition of these gates, which was very valuable. They were the ones who knew controlled the money and made the allocation decisions, so they knew what information they based these decisions on. At each gate, projects were either killed or prioritized against other projects, based on schedule, ROI, and probability of success.

This might not be helpful at all, but I'd suggest you make the administration a part of these discussions when you're defining the gates. You not only need their expertise, but you also need their buy-in. They're much more likely to support you if they contributed to the plan you're executing. You'll probably have to walk through and define that cone of uncertainty and ask what kind of information they need to promote something through that cone (i.e. from an idea/concept to a definition to a POC/design). If you can show how poorly defined projects can be eliminated early on and save money while still promoting new ideas, maybe they'll understand why you can't have 100% of the details "up front." Perhaps they misunderstand the entire life cycle of your projects, and they think that "up front" only occurs at the final stage.
...
1 reply by January Benjamin
May 29, 2019 1:27 PM
January Benjamin
...
Thank you. Good ideas. I did work with them to educate them on the Stage Gate process and they contributed ideas and feedback. Project management is a completely new concept, so you are right they are not clear on the life cycle of projects yet.
Network:28



May 29, 2019 1:22 PM
Replying to Wade Harshman
...
One place I worked used very strict stage gates, but we didn't have a formula. We had a general R&D process and defined what we needed to know at each gate. The executives were involved in the definition of these gates, which was very valuable. They were the ones who knew controlled the money and made the allocation decisions, so they knew what information they based these decisions on. At each gate, projects were either killed or prioritized against other projects, based on schedule, ROI, and probability of success.

This might not be helpful at all, but I'd suggest you make the administration a part of these discussions when you're defining the gates. You not only need their expertise, but you also need their buy-in. They're much more likely to support you if they contributed to the plan you're executing. You'll probably have to walk through and define that cone of uncertainty and ask what kind of information they need to promote something through that cone (i.e. from an idea/concept to a definition to a POC/design). If you can show how poorly defined projects can be eliminated early on and save money while still promoting new ideas, maybe they'll understand why you can't have 100% of the details "up front." Perhaps they misunderstand the entire life cycle of your projects, and they think that "up front" only occurs at the final stage.
Thank you. Good ideas. I did work with them to educate them on the Stage Gate process and they contributed ideas and feedback. Project management is a completely new concept, so you are right they are not clear on the life cycle of projects yet.
Network:307



There is a classic "cost influence model" you could use. It has a lot of adaptations such as the MacLeamy curve.

They show that over time, the ability to change goes down while the cost to change increases rapidly. Other versions include more lines for things like "total project knowledge", "uncertainty" and "total project cost" but they all describe about the same thing. Early on you have the most impact on cost and much of your cost gets baked-in early on.

Changes are inevitable, but you want to make the major ones early on. This doesn't mean you need to know everything up front. I don't know what your gates are, but typically in the conceptual design phase, you have a relative rough idea of the solution with a number of problems to resolve. In the following preliminary design phase, you are working out the major bugs. This stage is critical. Then comes detail design where the effort goes into refining the solution. Since the level of detail has grown, the rework grows quickly. If fabrication is involved, once you start the cost to change again increases significantly.

Cost curves differ by industry, but often gates are aligned to where cost curve changes slope significantly. The goal of the PM is not to know everything at the beginning, but to know when information must be known during the development lifecycle.

Please login or join to reply

Content ID:
ADVERTISEMENTS
ADVERTISEMENT

Sponsors