Project Management Central

Please login or join to subscribe to this thread

Topics: Agile, IT Project Management, Scrum
What is the optimal duration for an Agile project?
Network:929



For traditional (planned) approach there are recommendations for the size and duration of IT projects depending on the type of work (new development/enhancements/infrastructure).
Unfortunately for Agile projects there is no reliable data on the relationship between project size/duration and the rate of success of efficiency.

In you experience what is the optimal range for an Agile project in sprints/months, $ and Team size..
Sort By:
Network:325



It depends on the individual project and project type. For example, a classical SAP Template Build project should not exceed 1.5 years max. as you have normally business people assigned to the project and most of them are very eager to go back to their business line activities (especially from the career perspective). In this context the duration is independent from adaptive or predective prpject life cycle.
In a scaled environment individual team sizes should not exceed 12 max., better is 7-9 (but not always possible). In SAP projects I found 4 week sprints more suitable for the teams than 2 weeks sprints (but also in this regards there might be different experiences).
Network:1768



For all type of projects, using any type of approach, the optimal duration is the opportunity window the organization is trying to get with the solution to be created by the project.
Network:2127



Optimal should be based on the team's maturity, velocity, and their ability to continually deliver valuable, releasable, working software. Optimum should not be judged simply on timeframe and dollars.
Network:157



Agree with Sergio.

Duration for MVP/MMP varies.
Network:929



Thank you all, I wasn't looking for it depend on.... For planned projects there is enough data to determine the optimal range in terms of size (duration, cost, team). I was wandering if anyone did a similar study in Agile projects.
@Craig, I believe that optimum should extend to enterprise level. At the team level is practically irrelevant because the team composition can, and in Agile must, change. Using proper metrics it can be determined what project are the most efficient. Too small as well as too large project are less efficient.
@Sergio, in Agile the opportunity window is addressed by frequent prod releases. Even in planned approach a project can have multiple prod releases but in Agile it has to have more than one release, preferably at the end of each Sprint.
...
1 reply by Sergio Luis Conte
Feb 04, 2019 6:14 AM
Sergio Luis Conte
...
Just a first comment to clarify my point. When organizations need to transform or need to create a product/service/result to take advantage of an opportunity there is a time to do that. That is what I call opportunity window. No matter the way you use to get it the time is limited by lot of factors mainly because there are others outside there that want to get the opportunity too.
Network:145



A good question, but I do not think that having an absolute answer for the question, we need more statistically survey on this topic. As Chaos's Annual Report.
Network:1768



Feb 03, 2019 9:43 PM
Replying to Stelian ROMAN
...
Thank you all, I wasn't looking for it depend on.... For planned projects there is enough data to determine the optimal range in terms of size (duration, cost, team). I was wandering if anyone did a similar study in Agile projects.
@Craig, I believe that optimum should extend to enterprise level. At the team level is practically irrelevant because the team composition can, and in Agile must, change. Using proper metrics it can be determined what project are the most efficient. Too small as well as too large project are less efficient.
@Sergio, in Agile the opportunity window is addressed by frequent prod releases. Even in planned approach a project can have multiple prod releases but in Agile it has to have more than one release, preferably at the end of each Sprint.
Just a first comment to clarify my point. When organizations need to transform or need to create a product/service/result to take advantage of an opportunity there is a time to do that. That is what I call opportunity window. No matter the way you use to get it the time is limited by lot of factors mainly because there are others outside there that want to get the opportunity too.
Network:200



Stelian -

In my experience it usually goes for minimum 6 months and extended if need arises. Sprint is of 2 weeks duration and work is done on Scrum framework. But in case of projects where Kanban is being used, duration is sometimes indefinite as we keep working on continuous improvement items.

Team size of each Scrum/Kanban teams is mostly 9 but the total team size of all Scrum teams working on a Product would vary from 20 to 50 and in Scaled Agile ARTs they are nearly 90.

I also agree with Sergio, Andrew & Peter.
...
1 reply by Stelian ROMAN
Feb 06, 2019 5:10 PM
Stelian ROMAN
...
Thank you. Minimum 6 months seems logical for a project and it is close to the data that I have for planned approach projects. We did a study over 15 years with very useful and interesting findings but I left that organisation in 2010 and the project was closed soon after. We had very limited data for Agile projects because at that time it was less than 5% compared with project using planned approach.
Interesting also the comment on kanban. Projects are characterised by determined duration, it looks like kanban is good for BaU, logical conclusion considering that it originated in manufacturing where the same product was delivered over many years.
Network:929



Feb 05, 2019 4:00 AM
Replying to Girija Ramakrishnan
...
Stelian -

In my experience it usually goes for minimum 6 months and extended if need arises. Sprint is of 2 weeks duration and work is done on Scrum framework. But in case of projects where Kanban is being used, duration is sometimes indefinite as we keep working on continuous improvement items.

Team size of each Scrum/Kanban teams is mostly 9 but the total team size of all Scrum teams working on a Product would vary from 20 to 50 and in Scaled Agile ARTs they are nearly 90.

I also agree with Sergio, Andrew & Peter.
Thank you. Minimum 6 months seems logical for a project and it is close to the data that I have for planned approach projects. We did a study over 15 years with very useful and interesting findings but I left that organisation in 2010 and the project was closed soon after. We had very limited data for Agile projects because at that time it was less than 5% compared with project using planned approach.
Interesting also the comment on kanban. Projects are characterised by determined duration, it looks like kanban is good for BaU, logical conclusion considering that it originated in manufacturing where the same product was delivered over many years.
Network:929



I also agree with Sergio and Andrew but I was not looking for 'it depends'. Of course that the size depends on the scope, environment, complexity blah blah blah. That's common sense, I was looking for practical experience.

Combining yours and Peter's answers it looks like the optimal duration is between 6 and not more than 1.5 years. Probably for an Agile project 1 year should be a more realistic target.

This is not an academic exercise. Having this kind of benchmark it helps finding funds for projects.

I've seen 'projects' with 3 sprints of 2 weeks and projects that took 3 years to complete. Neither were efficient use of resources.

The 5 to 9 people size of the team didn't originated in Scrum or Agile. It is the result of scientific research and it doesn't apply to projects only. It is the optimum size that can be managed by a person.

Please login or join to reply

Content ID:
ADVERTISEMENTS

"Anyone who is capable of getting themselves made president should on no account be allowed to do the job."

- Douglas Adams

ADVERTISEMENT

Sponsors