Luis Gonçalves

Welcome to my site

Getting Value out of Agile Retrospectives: “Starfish”

title_page

Starfish exercise

In this post, I will explain the exercise known as “Starfish”. This exercise can be found in the book Getting Value out of Agile Retrospectives, a book written by me and Ben Linders with a foreword by Esther Derby. The book can be downloaded for free at LeanPub.com or InfoQ.com; please download it and share it with your colleagues.

What you can expect to get out of this technique:

The Starfish exercise is based on an evolution of the typical three questions that are used for retrospectives:

  • What went well?
  • What did not go so well?
  • What will be improved?

Instead of the typical three questions, the starfish exercise contains a circle with five words:

  • Stop – These are the activities that do not bring value to a team or to a customer. Activities that bring waste into the process.
  • Less – These are activities where the effort required to perform such activities is much smaller than the benefit. (Or activities that were brought into the team in the past but did not show any overall improvements to a process.)
  • Keep – Usually these are good activities or practices that team members want to keep. These activities are already being applied.
  • More – Activities on which a team should focus more and/or perform more often. For example, many teams tell me how pair programming is good, yet they do not do it each time they should.
  • Start – Activities or ideas that a team wants to bring into the game.

With the starfish exercise, teams can get a good overall picture of what’s going on within the team, what is working and what is not. They can get an overview covering both failed and successful work experiences in the past. In my personal opinion, I think this is a great evolution of the typical three questions.

When you would use this technique:

I believe the starfish exercise is quite simple and does not require any special occasion. Although, it might be interesting to apply it to situations wherein a team went through several ups and downs during the iteration. This technique reveals all the good things and less positive things achieved by a team. Therefore, this might be a good tool with which to make a summary of the sprint.

Starfish is suitable for any team; it does not require any specific level of maturity.

How to do it

Star Fish - Retrospective The starfish exercise is quite simple: First, we draw something like what is shown on the picture above in a flip chart. One of the beauties of this exercise is the fact that collocation of a team is not mandatory. You can use, for example, tools like Lino to apply the exercise to non-collocated teams. This tool allows us to do everything that we need to in order to run this exercise.

After placing the picture on a flip chart, it´s good to start a brainstorming session with your team, allowing them to dump their ideas in the “Stop” area. After that, give 2-3 minutes to each person to read out loud their “stop” ideas. Afterwards, spend 10 minutes on a short discussion to see if everyone is aligned.

Repeat the exercise for each of the different parts: “Less”, “Keep” and “More”.

For the “Start” part, add one extra step and use the Toyota approach, choosing one single topic to “Start”. I would ask for votes to see what the most important topic is, and that is the one the team should start with. After selecting the topic, design a small strategy to make sure a topic is well-implemented. This strategy might include responsible persons, the due date, and the most important success criteria. In order to know if the implementation was successful or not, we must have success criteria outlined.

I would like to highlight the fact that the theme that is chosen in the “Start” part does not need to be a new topic for a team, it can be an improvement on something that is not working well within the team.

Another important thing that is worth mentioning is the order of different “words” in the circle. I really like to start with: “Stop”, “Less”, “Keep”, “More” and finish with “Start”. I think this has a big impact. Starting with negative topics and progressing little by little towards the positive ones will help the team to end the retrospective with a much more positive feeling than if they did it in a random order.

I honestly think the starfish exercise is both pleasant and effective, but I would love to get your feedback. Please comment on this and let me know your opinion.

On my next blog post I explain how to use the “Value Stream Mapping” exercise to run a retrospective.

Did you like this post? Get a sample of the book subscribing my newsletter below and follow me on twitter: .

title_page




11 thoughts on “Getting Value out of Agile Retrospectives: “Starfish””

  1. mick maguire says:

    Hi Luis, thanks for taking the time to think this up and put it out there. I’m not sure the teams I have coached would find value in having more categories – e.g I would specultate that the difference between less and stop, and similarly between more and keep may not seem to add value for some teams.

    However, I’ll certainly take the opportunity to drop it into some of those that I am working with now. It is always good to have something fresh and new, even if its a variant on something old. Every team is different so this may provide a certain spark for a certain team.

    Thanks again…. Ready to have my speculation debunked :)

    1. Hi Mick,

      Good point the one that you brought :) but I still think is useful to separate them :)

      Below I am quoting what I wrote above, I guess isolating the definitions will help to see the differences :) :)

      Stop – These are the activities that do not bring value to a team or to a customer. Activities that bring waste into the process.

      Less – These are activities where an effort required to perform such activities is much smaller than a benefit. Or the activities that were brought into the team in past but did not show any overall improvements to a process.

      Keep – Usually these are good activities or practices that team members want to keep. These activities are already being applied.

      More – Activities on which a team should focus better, perform more often. For example, many teams tell me how pair programming is good,yet they do not do it every time they should.

      But Thank you so much for your comments :)

      Luis

  2. joejv says:

    This appears to be a variation of the “start”, “stop”, “continue” exercise. Not sure we need the additional categories.

    1. Hi, I guess you can create many many different variations :)

      I see your proposal quite close to the normal three typical questions:

      What went wrong?
      When went right?
      What will we change?

      And I see this approach as a bit boring :) Thats why I prefer something different, but then again this is my personal feeling :):).

      Thanks for your comment.

      Luis

  3. rik sagar says:

    I am a Certified Scrum Master, however, I am also a Certified Scuba Dive Master.

    As the former, looks good. As the latter, I must point out that “Star Fish” is an inaccurate term. Technically, you should call them “Sea Stars”.

    Sorry to be so pedantic, but as MASTERs it’s what we must do.
    http://animals.nationalgeographic.com/animals/invertebrates/starfish/

    Regards,
    Rik.

    1. I would say exactly the other way around :) As MASTERS we must be humble and simple :). But I liked your sense of humour :)

  4. Dodger Arnett - NZ says:

    I think we loose track of what a retrospective is for; it’s all about constant improvement. Big change we implemented as a team was to make the retros real and sizing the constant improvement tasks to something the team can work on within the next sprint.

    To achieve this we’ve simplified the retros so that its Start/Stop. It makes things clearer, quicker and allows conversations to focus on what the team can influence.

    Our reasoning for only Start/Stop was;
    - There are no real ‘Do Less’ because no one wants to do less bad stuff, they want to do no bad stuff.
    - Little benefit ‘Do More’ because its a given that if we doing it now and we don’t want to stop doing it then its good stuff and we should do more of it.

    One last thing; we all start with the ‘Stop’ when we list out the retro items. Even at the start of this thread it was “What went well, …,…” but the bullet point list starts ‘Stop’, ‘Less’,….

  5. A reblogué ceci sur Agillys and commented:
    Add your thoughts here… (optional)

  6. Arvind says:

    Good one.. Thanks for sharing a different perspective

  7. cnutsr says:

    Retrospectives evolve with every team, I generally introduce them to Start, Stop & Continue which has worked for most of the teams, “More & less” needs to be tried…I believe it is the responsibility of facilitator, coach & team to make this process flexible for everyone involved and drive clear action items to closure for giving the team all it requires to deliver the software…:)

  8. Azra Irshad Rabbani says:

    It seems that most of the people agreed with the concept of stop, start and continue (which is quite traditional as well) but I agree with Luis and i think there can be somethings which can fall under Less and More as well… e.g. there can be some practices which cannot become a part of our process due to some traditions, some constraints (resources, time etc.) but they produce better outcome whenever adopted at the right time and right place as Luis gave example of Pair programming.

    ‘More’ are the practices which we would encourage but can’t make them part of the process or we can say that they cannot be as useful in all situations so we would like to keep them ‘on demand basis’.
    So considering everything we can have this mutual understanding that whenever it seems required a person shouldn’t hesitate but go for it.

    Likewise ‘Less’ are the practices which we neither encourage nor remove.
    These are the practices which we need to do in some particular situations but they are the things which should declare as ‘should be avoided until and unless there is a severe need of it’.

Leave a Reply

Your email address will not be published.

*

Facebook

Get the Facebook Likebox Slider Pro for WordPress