Reflections on a new retrospective

One of the most important things to my team and myself is to learn from our experiences hence I think reflecting over our actions is essential for creating such an opportunity.

I don’t get to facilitate retrospectives that often since I’m not really a part of any project or any team on daily basis. But I love to do it whenever I get the chance.

I like trying out new methods in my meetings and adjusting them trying to increase the personal  ROTI for each meeting member( I find this method very efficient). To be fair though my goal is at minimum 4.0 as an average. Note that I have been using the scale 1-5 since I am not sure if it makes a difference from the 0 – 4 scale.

My team has grown quite lot during the last year and lately my ordinary type of retrospective had gotten a lower and lower ROTI average. The main reason have been the lack of discussion and the time for individual notes written and communicated. With 19 people today it was not a very efficient way of facilitating a retrospective any more.

Today I tried something new during a retrospective we had for a three day testing activity. I accidentally stumbled on a tweet about Dialogue sheets. If you haven’t heard about the technique before you can read about it here. I thought it sounded like a really fun way to facilitate a retrospective. It even gives the possibility to remove the facilitator role. But facilitating is something I really like so I decided to do at least some facilitation including preparation.

Sixteen people (excl. myself) participated. There were a few challenges with this:

  • Two of my team members are physically on another site.
  • The room was not really designed for this type of retrospective. It is a conference room with a long table for 16 people.

It was possible to break this long table in to three different tables so the second problem was easily solved. The first problem was supposed to be solved by video communication but of course the technology didn’t work. So the two guys on a different site were left in isolation to work on the dialogue sheet.

After the tables were moved around I asked the team members to group in two equally big groups and take place around the tables and start. Most of the team members had informed them selves about Dialogue sheets. Then I gave them 60 minutes to complete the Dialogue sheet. I left them managing them selves and tried to observe the two different groups.

I named the groups A, B and C where the latter was the one on a different site.

I had to time box the completion of the sheet to have room for comparing between the different groups. I chose not to include Group C when comparing the dialogue sheets since they would have to follow the discussion on a normal phone. From previous experience of discussions over the phone where many people are involved it gives very little ROTI for the few persons on the other side. I decided to catch up with Group C later.


During the meeting I did following observations of the two groups, Group A and Group B.

Group A

  • Some one in the group said “Let’s stand up and read the instructions”
  • The group stood quietly reading.
  • Then some one said “OK lets sit down again”.
  • The sheet was not place in the center of the table.
  • There was much less laughing then in group B.
  • There were mainly two people talking.
  • Little notes on the sheet itself
  • The group was ready after 40 minutes.
  • One person was writing on a paper.
  • Notes were handed to me after the time was up.
  • Mainly one person spoke when sharing what had been discussed during the 60 minutes.
  • Lower average ROTI then Group B

Group B

  • The person closest to the first point started reading the instructions out loud.
  • The discussions started almost immediately.
  • The sheet was placed in the center of the table.
  • The discussion were vivid and pretty loud.
  • Not everyone had a pen at start.
  • There was an intense discussion on the meaning of success.
  • There was a lot of laughter.
  • Most of the people in the group was speaking though I noticed that there were a few people speaking more than others.
  • Far more notes on the sheet.
  • When time was up the group continued writing on the sheet.
  • During sharing what had been discussed several people talked. There was one person who did most of the talking.
Dialogue sheets

Dialogue sheets

ROTI rating the personal time invested 

After a quick hand raise in the end of the meeting and after checking with Group C  we ended up with 4,4,4,5,4,4,4,3,3,2,4,4,4,4,4,2

Following up on the numbers:

5 – The method opened up the possibility for discussions. We eliminated the writing post-its part where every one sits quite and glances on the others writing post-its.

2 – The discussion sometimes felt constrained since we had to “follow”  the board. There was a good discussion on going but it was interrupted because some one mentioned that we were not supposed to talk about that subject. So the discussion ended and went back to something else. To change the rating perhaps facilitating the discussion would help. Or when we get a better grip of how the board is used.

2 – Personally I don’t have a lot to benefit from the retrospective since the result will not affect me (the purpose of the activity that we did the retro for is quite different for this person). I did like the setup though.

3 – The new method was good though it didn’t do any big difference from other meetings. It was difficult to focus on one question at a time. I like the Action plan. One thing negative thing is that some one can dominate the discussion. 

3 – There was more discussion then in previous meeting. Unfortunately there was no one leading the discussion. Once we had a great discussion on going some one noticed we were on the wrong topic and that it was something that shouldn’t be discussed now. Some one in the group had the idea that positive events was supposed to be written above the time line and negative events below and some people thought you could write it anywhere. A facilitator would have helped.


I am not going in to detail about my analysis of these observations and personal grading of the meeting. But I found the result of the different groups very interesting. I also wonder if the groups forming was a conscious action or an unconscious action. I think the difference in the groups depends on a few things:

  • the different personality types of each individual
  • and the amount of time people have known each
  • how well they know each other

In Group A the majority of the people have worked less than a year at the company and in Group B the majority of the people have been working for several years together.

Actions for next meeting:

For the next retrospective which will be in a month I am considering a few things that I might do. I haven’t yet decided. I also have a few reflections and a few decisions to make:

  • Should we keep the same groups next time ( It will probably no be the same any way. It might be that the same people can’t attend. I might decide this when I know who are attending next time)
  • Should the groups be smaller, maybe break in to 3 groups?
  • Should I let people decide in how to group within certain limits?
  • Should I try to facilitate? Though it will be difficult to facilitate several groups at the same time. If some one else in the group facilitates they won’t be part of the discussions.
  • Should I try to set up some more “rules” then what is stated on the board. For example decide on how to facilitate your own group. How to keep the discussion ongoing before starting.
  • Should I facilitate or just observe?

What I will do:

  • Get pens in different colors to make the board more colorful. It might increase the writing on the board.
  •  I’m keeping the facilitator role to facilitate the exchange of information in between the groups, to keep track of time and make sure we retro the retro (ROTI).
  • If I decide to observe, which is what I would prefer to do I need to define my method to make better observations. I will time box time spent on each team. I will choose a few specific things to observe.
  • During ROTI I will also ask the participants who rated the meeting balanced something that I have totally forgotten about.

9 thoughts on “Reflections on a new retrospective

  1. Very interesting observations. Deaking with a distributed group is always hard, and I think the only way would be with the help of tech.
    I really like dialogue sheets for the retrospect. I think it helps the discussions, without people being aware of it. When everyone is forced to read and take an active role, they tend to engage more. I don’t think you should facilitate a workshop with sheets. If you want to stay an observer, or even a participant then have someone else facilitate. Perhaps someone external.
    Don’t keep the groups. Mix people up as much as possible. Try to force them working with someone they don’t know very well. Sometimes people become more open and curious this way.
    Good luck!

    • Thank you Jakob for your comment!
      Next time I will not have a distributed group to deal with since the person will come to join us in Malmö. So that obsticle is removed for next time at least.
      How have you dealt with the fact that some people might dominate at discussion?
      How often do you have retrospectives?

      • Having people dominate a discussion is always a challenge. Depending on the group, silently writing post its, and then posting them together using affinity mapping ( could be a good way. I’ve also had success using “Locate strengths” (from Larsen’s Agile retrospectives, people interview each other in pairs about high points in the last iteration). Splitting people into pairs or smaller groups for part of the workshop usually helps.
        I find that another important thing is doing some kind of check in at the beginning, make everyone say something before the retrospect actually starts. Then it’s easier to keep talking during the meeting. I like doing the checkin from the core protocols (
        So, my take is rather than silence the people that talk a lot, make the others talk more.
        We do retrospectives after each iteration, which currently means every second week.

      • I’ve been considering using a check-in before the retrospect starts.
        It’s good to have some tools to use and I’ll probably decide what to do once we are on.
        Thank you for your tips!

  2. Happy to hear you have found something that makes you get out more from you retropective.
    At the company I work, some team uses something similar to the Dialogue sheets, Its the same concept but whe use a whiteboard as an timeline and post-it papers with different colors.

    Here in more in detail:
    1. Set the stage – Dra participants attention

    2. Gather data – Draw the sprint timeline on the whiteboard. Then everyone write on standard post-It papers what they have done. After ~5 min each participant puts them up one the timeline and explan it. The pupose with this is to refreash what has happened during the sprint.

    3. Generate insights – Each participant writes new post-it papers (~10min), but with colors, what is going well (green), what is not going so well (red) and new ideas to improvement (yellow). We also have a flower to reward when someone has done something really god (orange).Put them up on the white bord in colorised groups and discuss around it.

    4. Decide what to do: Pick 1 to 3 improvements or experiments that will make a difference for the team.

    5. Close the retrospective

    As you and other has mentioned in the comments, everyone got to speek, even thoose who normally is silent.

    // Pierre

    • Hi Pierre,
      The process you are describing is also a good process to use. This is the one I was using before. But as I wrote it doesn’t really work any more. My team is sometimes 19 people and the number of people is too high to have an efficient discussion. Having 19 people standing by the time line explaining what was most important to them takes a long time and there is no interaction really within the group. The dialogue sheet is almost the same but you sit together and discuss. The time line is also there. Like Jakob wrote it is also a possiblity to mix post-it writing with the dailogue sheet.
      Thank you for reading an commenting my post!

  3. Thanks for the write up, really interesting.

    I recognise some of the points you made. For example I’ve taken to carrying a set of Sharpies pens with me to lend to teams doing Dialogue Sheets because I find the colours and large writing add to the energy in the sheets.

    I also think the position of the sheet makes a real difference. When it is in the centre of the group it is more egalitarian. When a team sit on one sheet – and move the sheet or something – then the dynamics can lead to a few people being more active than others.

    As to the future…. I sometimes mix people up randomly between groups and sometimes try to group them by something in common, e.g. Testers and Analysts on one sheet, Developers on another; or the people who work on product X on one and product Y people on another. So I’d probably not use the same teams next time.

    I also find that the more people you have on a sheet the longer it takes. A team of 6 might need a bit more than an hour, while a team of 3 might finish early. (You account of one large team finishing early does sound unusual but then so does the way you describe two people dominating the conversation.)

    Again, thanks for trying the sheets and thanks for the write up.

    • Thank you for taking the time to read my write up and comment on it.
      The teams will actually not be the same next time since it is already known that some people can’t participate and others that couldn’t participate the last time will actually participate next time.
      So observing those groups will be interesting. What part do you take during the use of the Dialogues sheets? Are you facilitating the meeting such as you would move a sheet back to the centre of the group?
      I’m looking forward to use the Dialogue sheets again!

      • I’m rarely a part of a team these days – the perils of being a consultant – so I normally end up floating. You might call this being a facilitator but generally I find teams understand what is to be done. I brief them before we start and check up on them once or twice during the session but there is little else I find needs doing. Sometimes people ask about question wording, usually I try to avoid answering because some of the questions are deliberately open to interpretation.

        The only real act of facilitation comes at the end when there are multiple groups on multiple sheets. Then I ask someone from each group to summarise.

        Although I always say “No need to read the answer to every question, just give us the key points” almost everyone goes over every question.

        I will not the final “3 things” on a flip chart for each group, in the end there are normally there are more than 3 things in total so I might do a dot vote, or sometimes multiple groups have come up with the same thing so the top ones are obvious.

        In this format there isn’t really enough work for a facilitator – I sometimes go to Starbucks while the teams are working! A team member who was facilitating could join in with one group.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s