Agile Retrospective

Design Patterns

About me

Nico Riedmann
something @ somewhere

💻 riedmann.dev
📧 nico@riedmann.dev

Design Patterns

Photo by Ricardo Gomez Angel on Unsplash
Left to right, alex, Pallazo Poggi, Myotus (CC BY-SA 4.0), Ethan2039 (CC BY-SA 3.0) via Wikimedia Commons'

reusable solutions to common problems

Retrospectives

inspect and adapt methods and teamwork after an increment of work

Agile Retrospectives by Esther Derby & Diana Larsen

A Retro format pattern...

A Retro format pattern...

  1. Set the stage
  2. Gather Data
  3. Generate Insight
  4. Decide what to do
  5. Close

...you probably see a lot

  1. Good/Bad/Start/Stop
  2. Dot Vote
  3. Discuss and Decide on Action Items

reusable solution to

structure

not to

content

Patterns to Content

Patterns to Retrospectives

consider which
"common problem" you're solving

"common problem"

team

"common problem"

team + situation

no reusable solutions

to uncommon problems

😒

But sometimes teams are in common situations!

Let's take a look at some!

A large new team just formed from members of exisiting teams, most don't know each other, some have bad opinions of other teams from the past

  1. check-in - set the mood for open discussion
  2. sailboat - look forward, with room to address risks
  3. 5 whys in pairs - dig into why things could go well/wrong, and build 1:1 connection
  4. share
  5. appreciations - room to share something you appreciated about a team member in the retro or iteration,

A long runnnig team that get's along well had an unusually successful sprint

  1. reflect on the stories and happenings in the last iteration
  2. "If we had ruined the last sprint what would we have done?" - collect and compare to what went well
  3. decide on one thing to keep doing to keep having great sprints

Why? - Show the value of tailoring Retros to your team, while following an outline to achieve what you want - Link to design patterns as tools to resolve known problems What? - Design Patterns as tool to solve known problems - Purpose of retrospective - General structure - Some Patterns - sample situations - activity - team member tension - really successful sprint

First a little bit about myself...

Let's start with the basic concepts. Design Patterns. Most of you can probably think of some design patterns right? (audience engagement - expect Software Desgin patterns)

I was thinking more along the line of these Design Patterns. The concept was originally coined in architecture

defining reusable solutions to common problems. Which is exactly how we apply them in software, and I'd argue agile retrospectives.

So what is a retro for actually, the definition I like best is

a team gathering to inspect and adapt methods and teamwork after an increment of work. Which is how Derbe and Larsen define it in their "Agile Retrospectives" book, which I am a huge fan of and which defines a pattern for retros that you probably know

Which you probably see and use a lot in this form 2. Start/Stop/Continue 3. Dot Vote 4. Discuss and Decide on Action Items I'm not a huge fan of this exact "retro pattern", because

It is a reusable solution to the structure of a retro

but we often use it as reusable content as well.

In terms of content or activities, there's load of them, whith many great books and tools like retromat collecting them. And they're just ideas that worked for someone in the past, we may have better ideas to fit our teams problems.

So, we have a pattern to structure and lots of sources for activities, we're set, right?

Nope. Following a structure like the Agile Retrospectices one and filling it with random activies, does not make for a good retro either. When thinking of retrospective patterns,

consider which common problem you're actually solving.

conisder the team

and the situation they are facing at that moment.

And as these are very much individual to the moment and people, we have a hard time finding reusable solutions, if there are no common problems.

Luckily often times they are still "common problems" - it may be a need to foster collaboration in a 'storming' team or sprint goals that are never achieved. Knowing the problem we can look at all the options out there and choose the right "reusable solutions" that fit what we need.

<Interactive if there's time, else present> So let's take a look at two teams and situations that I made up and did not experience at some point or another...

haven't worked together low trust we're starting something new

What I would do: 1: - needs a check in, set the mood for personal discussion. Something setting communication rules (e.g. Focus On/Off) - Sailboat or good and bad future oriented format - 5 whys in subgroups to get to reasons, possibly with dot voting before if too many topics - Circle of Questions - going in a circle ask a question, next person ansers, then asks a question; to decide on ONE action and allow discussion on what and why of the action - Appreciations: Give room for telling other team members something you appreciated them do in the retro or iteration, no one has to speak. Why? New team, with some previous bad opions, I'd want to focus the retro on conversation giving room for people voicing their thoughts and opinions. Sailboat helps discuss the outlook for the future and possible worries, then allows discussion of how to overcome the bad, or make sure the good happens. End on positive personal closing activity. Likley takes 1.5h!

2. - likely doesn't need a check in, maybe just a quick "Describe your current mood in one word" - Reflect on every story in the sprint - did it go well or not, Why? - "If we had ruined the last sprint what would we have done?" - collect the "Bad Sprint" on one board, then collect the opposite of this on another (https://retromat.org/en/?id=74) - Now that we should have a decent idea of how a "good" sprint happens, decide on one SMART goal that helps make the next sprint good. - likely doesn't need a closing - good moment to gather feedback on the retro, e.g. 5-finger voting from 'waste of team' to 'super helpful' Why? Team is mature and performing, so focus on 'how we can we keep doing great' without too much format