Wanting to Implement Scrum, Manager asking for Project Plan

Last post 03:26 pm February 21, 2017
by Ashish Mathur
4 replies
Author
Messages
01:55 am February 15, 2017

Hi, 

I am wanting to implement Scrum for a new project. The timeline is 4-6 weeks to completion. My managers are very waterfall minded and want to see a project plan. 

In Scrum (and agile) a core tenant is responding to change > project planning and so I feel conflicted about what to do. When I told them how scrum works and suggested 10 3 day sprints, they suggested i list all the requirements i think need to be covered, assign times to them and them split them across the 10 sprints as I see fit. I have a bad feeling about this. I am unsure how to proceed. What do you think is the best response/solution in my case?

 

06:53 pm February 15, 2017

Do you have a Product Owner to work with, and who would value incremental deliveries at that cadence? Also, is the Development Team satisfied that they will be able to deliver a useful "Done" increment within such time-boxes?

09:54 pm February 15, 2017

You could propose to the managers to wait until, at the very least, a first sprint is completed. You can then support this proposition by arguing that Scrum is an empirical framework, and that experience (on previous sprints) is used to forecast the project's future progress.

If you have a Product Owner, then he or she can begin building a Product Backlog. When describing the Sprint Review, the Scrum Guide mentions, amongst other things, that "The Product Owner discusses the Product Backlog as it stands. He or she projects likely completion dates based on progress to date (if needed)". If you are the team's Scrum Master, you can therefore coach the Product Owner to make such projections empirically. There are many techniques to do this, and Scrum does not prescribe any particular technique, but a high level estimation and a measure of team velocity are two tools that, when combined, can give you some sort of a timeline.

Ian brings up a very important question with the length of the sprints. In particular, a 3 day sprint will most likely be quite noisy, and I am not convinced that, in the spirit of empiricism, it will be of much use to provide useful information towards forecasts for the following sprints.

05:05 pm February 17, 2017

This is classic constraining both the expected "deliverable" and "timeline". I am in the midst of coaching multiple teams through similar situations as the org moves from classic waterfall to agile scrum. The leaders are relying on what they are most comfortable with. Two things - is agile really the right practice for what they want? If the leadership is committed to agile, then you need to spend time with the PO and leaders on setting up the initial backlog and get agreement to what is minimum viable product for release. Do not even promise or commit to a date.

It is critical to show progress through the sprint reviews (and even showing them progress from any stories that are completed before the end of the sprint). I find that 5-day sprints are the best, especially with new teams. There is a lot of mindset and habits that need to get formed. The 5-days allows a frequent enough review point to allow for changes that can be incorporated into the next sprint.

10:09 am February 20, 2017

Another thought on this, should the organizations not move away from Project driven deliveries to Feature driven deliveries.

I see similar challenges where I'm currently, and as long as there is still a mindset that projects have to delivered, these issues will keep coming.

I know organizations which have completely re-structured themselves to adapt to the Agile way of working.