Geetali KheraSr. Project Manager| Macys Systems and TechnologiesAlpharetta, Ga, United States
We are spending too much time estimating projects in the conceptual phase. These estimates are being used to decide if the project will get funded or not, or part of scope will get funded, which is the minimum viable product.
Architect and the analyst write a document with scope items , and a system diagram, and teams are identified who are supposed to join a meeting to review. Then estimates are due one week later. But some problems are 1) that right people don’t always attend the review, 2) estimates drag and people go too deep into estimating whereas we are looking only for T shirt sizing’s, and 3) many times teams are missed for impacts. Writitng the document to revieweing it to getting estimates and finalizing on funding takes 3-4 months back to back. Each month we review at least 3-4 potential projects for providing estimates. It’s a process by which project finally gets funded. And many of these projects NEVER get fuded, but we invested time and effort to size these. Looking for ways to make this less costly, less time consuming, more effective, and still get estimates and fund the right projects.
Senior Projects Manager | Field & Marten AssociatesNew Westminster, British Columbia, Canada
We more or less do the same - We do conceptual design and ROM Estimates. In order to be more efficient you need to have historical data base from past projects. It will allow you to estimate project faster while your estimates are realistic. Saving Changes...
Geetali KheraSr. Project Manager| Macys Systems and TechnologiesAlpharetta, Ga, United States
Thanks Rami. Do you base off your final budgets on this ROM you collect? Saving Changes...
Senior Projects Manager | Field & Marten AssociatesNew Westminster, British Columbia, Canada
Of course not - We use ROM's to give the client a sense of how much it would cost and as you might be aware, the precision of ROM's is usually + or - 50% but with historical data base of similar projects, you can drag it down to + or - 25%.
Final Budget is based on the actual design and tenders. Saving Changes...
Thomas WalentaGlobal Project Economy ExpertHackenheim, Germany
Agree with Rami, this is an important point: conceptual estimates are needed to make a go/nogo decision or to prioritize a portfolio. This is often done without a delivey project manager. If there is a go decision and a charter, the project manager should develop the definitive budget he or she commits to. The conceptual budget should be considered a constraint for this process. Saving Changes...
Geetali, since you are only looking for T Shirt size estimates, to avoid problem #2 that you mention, I would recommend changing your approach on who all is involved during the sizing exercise. It should be done at Director/VP level (perhaps this is what Thomas alludes to) and you can decide whether it makes sense that a PM is involved at this time.
I think sticking more strictly to analogous estimation (finding 2-3 projects that are similar to the current prospect project) and restraining from going into bottom-up estimation seems to be key in your situation - because the accuracy isn't necessarily guaranteed even if bottom-up estimation is used at this point (since not all resources are available to make the right kind of bottom-up estimation).
Now this may look like a major change for your organization, so you may want to gather some data (number of projects estimated in the last 12 months, effort spent using current estimation approach, number of projects actually approved, project effort by switching to strict analogous approach etc.) before making your case. If you do make the switch, you should also track the improvement (or vice-versa) going forward
...
1 reply by Geetali Khera
Jun 27, 2016 10:22 AM
Geetali Khera
...
Thanks so much. I think this is a good approach. Would like to propose this in our organization.
Saving Changes...
Sergio Luis ConteHelping to create solutions for everyone| Worldwide based OrganizationsBuenos Aires, Argentina
Please let me say this: what is your objective to spend much time (while in fact, we have to define what "much time" means to you)? Accurancy? Precision? Low risk?. Forget about it. Estimation depends on two critical factors: information and time. You can manage the second one. But you can not manage the first one. So, what is critical to not forget is this: Barry Bohem´s Cone of Uncertainty. While it was created for software industry along the years it is used for any type of initiatives. Create your best gest with the time and information you have available AND DO NOT FORGET to apply "Cone of Uncertainty" directives to your estimations.
...
1 reply by Geetali Khera
Jun 27, 2016 11:48 AM
Geetali Khera
...
Thanks. I need to read up on this "cone of uncertainty" . Will do :)
Saving Changes...
Geetali KheraSr. Project Manager| Macys Systems and TechnologiesAlpharetta, Ga, United States
Jun 24, 2016 2:19 PM
Replying to Samuel Vaddi
...
Geetali, since you are only looking for T Shirt size estimates, to avoid problem #2 that you mention, I would recommend changing your approach on who all is involved during the sizing exercise. It should be done at Director/VP level (perhaps this is what Thomas alludes to) and you can decide whether it makes sense that a PM is involved at this time.
I think sticking more strictly to analogous estimation (finding 2-3 projects that are similar to the current prospect project) and restraining from going into bottom-up estimation seems to be key in your situation - because the accuracy isn't necessarily guaranteed even if bottom-up estimation is used at this point (since not all resources are available to make the right kind of bottom-up estimation).
Now this may look like a major change for your organization, so you may want to gather some data (number of projects estimated in the last 12 months, effort spent using current estimation approach, number of projects actually approved, project effort by switching to strict analogous approach etc.) before making your case. If you do make the switch, you should also track the improvement (or vice-versa) going forward
Thanks so much. I think this is a good approach. Would like to propose this in our organization. Saving Changes...
Stéphane ParentSelf Employed / Semi-retired| Leader MakerPrince Edward Island, Canada
I'm with Samuel: avoid bottom-up estimating until the project planning stage. Anything before planning should use either analogous or parametric estimations. Saving Changes...
Geetali KheraSr. Project Manager| Macys Systems and TechnologiesAlpharetta, Ga, United States
Thanks all for taking the time and interest to provide your inputs. I have learned from each and every answer provided here. This helps me. I am glad to be part of this network that I recently joined. Saving Changes...
Geetali KheraSr. Project Manager| Macys Systems and TechnologiesAlpharetta, Ga, United States
Jun 24, 2016 3:02 PM
Replying to Sergio Luis Conte
...
Please let me say this: what is your objective to spend much time (while in fact, we have to define what "much time" means to you)? Accurancy? Precision? Low risk?. Forget about it. Estimation depends on two critical factors: information and time. You can manage the second one. But you can not manage the first one. So, what is critical to not forget is this: Barry Bohem´s Cone of Uncertainty. While it was created for software industry along the years it is used for any type of initiatives. Create your best gest with the time and information you have available AND DO NOT FORGET to apply "Cone of Uncertainty" directives to your estimations.
Thanks. I need to read up on this "cone of uncertainty" . Will do :) Saving Changes...
Community Champion
