Friday, February 4, 2011

How we do Sprint Planning 2

Many people think Sprint Planning 2 (the second half of the Sprint Planning meeting) is just about creating and estimating the list of tasks for the selected product backlog. This cannot be a bigger over-simplification! Here is simple approach to conducting an effective Sprint Planning 2 Meeting.

In Sprint Planning 1, the Implementation Team and the Product Owner negotiated which "stories" would be implemented in the coming sprint. The team made sure it understood the stories, in particular the acceptance criteria (I recommend agreeing on 'How to Demo').

During Sprint Planning 2, the Implementation Team must figure out how to solve the problem it took on in Sprint Planning 1. This consists of two parts:
  1. A solution concept - a design, architecture or whatever, which explains how the problem is to be solved/feature is to be realized.
  2. A list of tasks - what steps must the team do to get each selected backlog item to the state 'done'.
The goal is not an absolutely perfect design or task planning. It's about getting a clear enough concept that the team can start work.

In my coaching, I have found it effective when the team members work in pairs in a solve-present-solve-present sequence. They brainstorm the technical concept, then present it to the team. Depending on the feedback, they may need to enhance, modify or possibly even rethink the concept. Again working in pairs, they then create task cards, showing the route to "done." Finally, each pair presents these tasks to the entire team.

Let us assume the team has committed to 6 stories, that there are 6 team members, and that the team is doing 2 week sprints. So the Sprint Planning 2 is time-boxed to 2 hours. The team forms 3 pairs and each pair works on two stories. So the agenda for the meeting is as follows:

Agenda for Sprint Planning 2

14.00 - 14.05 Start - pair off and divide the stories among the pairs.
14.05 - 14.35 Concept - each pair brainstorms on the technical concept (producing a diagram, text or something which can be presented to the rest of the team)
14.35 - 15.05 Present - each pair presents their solutions to rest if the team. The presentation and Q&A time for each story is time-boxed to a total 5 Minutes (30 Minutes / 6 Stories )
15.05 - 15.35 Tasks - each pair creates a set of tasks to get their stories to 'done'. Each task represents a goal for the day. The tasks are not estimated, but should be completable in one working day or less. This saves a lot of effort!
15.35 - 16.00 Present - again, each pair presents the tasks needed to get their stories to done. In the course of discussion, they may discover missing or unnecessary tasks. Each story is time-boxed to 4 Minutes (25 Minutes / 6 Stories ).

This approach ensures that at least 2 people have thought intensively about each story. It gives structure to the meeting, so the team can accomplish the meeting within the time-box. And by presenting and discussing the solution within the whole team, the rest of the team is much better positioned to help out during the execution of the sprint.

6 comments:

Fabrice Aimetti said...

Hello Peter,

I've translated your interesting post in french :
Comment réaliser le Sprint Planning 2

Regards, Fabrice

Peter said...

Merci Beaucoup, Fabrice!

Lukas said...

Hi Peter. I consider the sprint planning 2 both difficult and crucial, so I appreciate your advice on this.
I wonder though whether it is really possible to present, ask&answer questions about a solution for a story in 5 minutes, let alone to present the tasks for a story, discuss them and discover missing/unnecessary ones
in 4 minutes.
Lukas

Peter said...

Hi Lukas,

Thanks!

The short answer is: Of course you can! Just set and stick to the time box. When the time is up, the time is up.

The goal of the Sprint Planning 2 is not an absolutely perfect design or task planning. It's about getting a clear enough concept about how to move forward that your team can start.

This approach is somewhere between an individually produced concept and planning poker. By getting the relatively large and diverse team thinking about the story, you get people with differing perspectives to look at the problem and provide feedback.

I would suggest you try it and see how it works (and post a comment here to share your experience)! Sometimes trying is more effective than discussing!

Lukas said...

ok, thanks, i will.
Lukas

Dov TSAL SELA said...

I like the idea of pair analysis! Thanks!