Project Management Central

Please login or join to subscribe to this thread

Topics: IT Project Management, PM in Academia, Schedule Management
Comparing "Planned Completion" and "Actual Completion" Durations
Network:17

Hi Everyone,

I have 100 small size projects with 7 steps (build, test, implementation etc.) and there are 15 various finish-to-start tasks until the PIR. Each project has a duration template and tasks can get pushed back, moved forward sometimes, and take longer than planned initially..
What I want is to see when and in which step those milestones were, and how much longer it took to complete the tasks and the whole Project than it should have.
Would analyzing the variance between "Planned Completion" and "Actual Completion" help me? What should I do to make this comparison correctly and easily? Because these kind of empirical studies are rare, and I am having problems with finding the right methodology in my case :(

Any ideas are welcome!- Thanks
Sort By:
Network:1510



Ezlena -

you can capture the data you need in a traditional scheduling tool like MS Project, or you could use a work board-oriented solution (e.g. JIRA) and report on cycle time by phase and/or overall.

Kiron
Network:69



I'm not very clear on what you're asking. What are you trying to get out of this? Is it to refine your plan to become more accurate over time? E.g. if the tasks are typically taking 10% longer than you want to focus on reducing this and/or reflecting this in the remaining tasks?

Are you also concerned about effort?
...
1 reply by Ezlena Lahmacunka
Dec 11, 2018 4:43 PM
Ezlena Lahmacunka
...
Thank you for your reply.
I 'd like to find the tasks which require more efficient PM performance in terms of managing time, for a better understanding on time-scope trade-offs .
Network:17

Dec 11, 2018 3:48 PM
Replying to Robert Neil Wood
...
I'm not very clear on what you're asking. What are you trying to get out of this? Is it to refine your plan to become more accurate over time? E.g. if the tasks are typically taking 10% longer than you want to focus on reducing this and/or reflecting this in the remaining tasks?

Are you also concerned about effort?
Thank you for your reply.
I 'd like to find the tasks which require more efficient PM performance in terms of managing time, for a better understanding on time-scope trade-offs .
...
1 reply by Robert Neil Wood
Dec 11, 2018 10:09 PM
Robert Neil Wood
...
I think you're saying: you want to identify those tasks that are generally taking longer than planned and focus on reducing them to your planned values. Is that right?

If that's the case, then yes, looking at the difference between planned and actual is useful. However, as you probably know, the difference alone won't tell you why but it's a good pointer to where you should look. Lot's of possible reasons why tasks could take longer (or shorter) than planned, including incorrect estimates, resource experience, impact of other priorities, differences in task details, impact of dependencies, vacations, etc.
Network:300



Analyzing the difference between planned and actual will give you a variance on how far off the estimates are from the actual completion if what you want is to evaluate your estimating ability.

If you want to focus on process improvement rather than estimating accuracy, then you should look at both the overall process time, and the individual steps by calculating the variance for each. You can calculate both the mean and the standard deviation using Excel functions or they're easy to calculate manually as well.

The tasks with the largest mean value are often where you have the most leverage to improve (with some exceptions like parts of the task are outside your control). The variance in the lengths for each task will give you better information on where standardization could yield the most improvement. The total variance will give you some indication of how much opportunity you have for the entire process. That is useful also as sometime spending more time on one step might reduce time in other steps so the total can provide value in addition to individual pieces so just trying to reduce the time of one step might not be best for the overall goal.
...
1 reply by Ezlena Lahmacunka
Dec 12, 2018 8:31 AM
Ezlena Lahmacunka
...
Thank you so much for the brief explanation. This is exactly what I am looking for as an approach. I will test all these steps, and will see how the results turn out :)
Network:69



Dec 11, 2018 4:43 PM
Replying to Ezlena Lahmacunka
...
Thank you for your reply.
I 'd like to find the tasks which require more efficient PM performance in terms of managing time, for a better understanding on time-scope trade-offs .
I think you're saying: you want to identify those tasks that are generally taking longer than planned and focus on reducing them to your planned values. Is that right?

If that's the case, then yes, looking at the difference between planned and actual is useful. However, as you probably know, the difference alone won't tell you why but it's a good pointer to where you should look. Lot's of possible reasons why tasks could take longer (or shorter) than planned, including incorrect estimates, resource experience, impact of other priorities, differences in task details, impact of dependencies, vacations, etc.
...
1 reply by Ezlena Lahmacunka
Dec 12, 2018 8:15 AM
Ezlena Lahmacunka
...
Thank you so much for your quick response. I will bear in mind.
Network:2265



If I understand correctly, could simply use MS project (or excel for that matter, but I find MS Project easier). Set up the project with all tasks and planned durations, then set a baseline. Add columns to track actuals (start, finish, variance). Leverage the reporting tools, and there you go!

Lots of videos on Youtube for tutorials.
Network:23762



I would say that review the concept of Value Added technique. try to match your needs with the tools and techniques you have.
Using applications like MS Project, Primavera, etc. can help you with calculations, however, you need to be familiar with the concepts as well.
Network:17

Dec 11, 2018 10:09 PM
Replying to Robert Neil Wood
...
I think you're saying: you want to identify those tasks that are generally taking longer than planned and focus on reducing them to your planned values. Is that right?

If that's the case, then yes, looking at the difference between planned and actual is useful. However, as you probably know, the difference alone won't tell you why but it's a good pointer to where you should look. Lot's of possible reasons why tasks could take longer (or shorter) than planned, including incorrect estimates, resource experience, impact of other priorities, differences in task details, impact of dependencies, vacations, etc.
Thank you so much for your quick response. I will bear in mind.
Network:17

Dec 11, 2018 4:56 PM
Replying to Keith Novak
...
Analyzing the difference between planned and actual will give you a variance on how far off the estimates are from the actual completion if what you want is to evaluate your estimating ability.

If you want to focus on process improvement rather than estimating accuracy, then you should look at both the overall process time, and the individual steps by calculating the variance for each. You can calculate both the mean and the standard deviation using Excel functions or they're easy to calculate manually as well.

The tasks with the largest mean value are often where you have the most leverage to improve (with some exceptions like parts of the task are outside your control). The variance in the lengths for each task will give you better information on where standardization could yield the most improvement. The total variance will give you some indication of how much opportunity you have for the entire process. That is useful also as sometime spending more time on one step might reduce time in other steps so the total can provide value in addition to individual pieces so just trying to reduce the time of one step might not be best for the overall goal.
Thank you so much for the brief explanation. This is exactly what I am looking for as an approach. I will test all these steps, and will see how the results turn out :)
Network:300



Glad I could help. With 100 projects you have a nice sample size for meaningful statistics. If you're looking for other related methodology on how to use the numbers you have, 6 Sigma sources are very good as it's basically how to use statistics to improve processes.

Please login or join to reply

Content ID:
ADVERTISEMENTS

"There is one way to find out if a man is honest: Ask him! If he says yes, you know he's crooked."

- Groucho Marx

ADVERTISEMENT

Sponsors