Planning your next agile retrospective? Start with a random plan, change it to fit the team's situation, print it and share the URL. Or browse around for new ideas!

Is this your first retrospective? Start here!
Are you running your retrospectives with Miro? Create prettier boards faster with the giant Retromat Miroboard Mega Template!

Check out the Mega Template
Plan-ID:
Replaced by JS

Check In - Quick Question (#3)

Ask one question that each participant answers in turn
Source: Agile Retrospectives
In round-robin each participant answers the same question (unless they say 'I pass'). Sample questions:
  • In one word - What do you need from this retrospective?
    When someone answers something that alerts you such as "help" or "protection", you have to react to that e. g. with "Is there something we can do right now to help?" oder "What kind of protection?"
  • What's something that caused problems last iteration?
  • If you could change one thing about the last iteration what would it be?

Avoid evaluating comments such as 'Great'. 'Thanks' is okay.

Quartering - Identify boring stories (#64)

Categorize stories in 2 dimensions to identify boring ones
Source: Wayne D. Grant
Draw a big square and divide it into 2 columns. Label them 'Interesting' and 'Dull'. Let the team write down everything they did last iteration on stickies and put it into the appropriate column. Have them add a rough estimate of how long it took on each of their own stickies.
Now add a horizontal line so that your square has 4 quadrants. Label the top row 'Short' (took hours) and the bottom row 'Long' (took days). Rearrange the stickies in each column.
The long and dull stories are now nicely grouped to 'attack' in subsequent phases.

(Splitting the assessment into several steps, improves focus. You can adapt Quartering for lots of other 2-dimensional categorizations.)

Speed Dating (#26)

Each team member explores one topic in depth in a series of 1:1 talks
Source: Thorsten Kalnin
Each participant writes down one topic they want to explore, i.e. something they'd like to change. Then form pairs and spread across the room. Each pair discusses both topics and ponders possible actions - 5 minutes per participant (topic) - one after the other. After 10 minutes the pairs break up to form new pairs. Continue until everyone has talked to everyone else.
If the group has an odd number of members, the facilitator is part of a pair but the partner gets all 10 minutes for their topic.

(The way to make all the pairings come out even is the following: Form two rows facing each other. This is pairing one. Pick one person. This person's position is fixed the entire time. Everybody else rotates one spot after each pairing.)

Planning Poker Voting (#99)

Use your Planning Poker cards for un-influenced voting
Source: Andreas Ratsch
If you've got very influential and / or shy team members you can re-use Planning Poker cards to vote simultaneously:

Write all suggested actions on sticky notes and put them onto a wall. Hand out an ordered deck of Planning Poker cards to each participant. Count the proposals and remove that many cards from the back of the card decks. If you've got 5 suggestions you might have cards '1', '2', '3', '5', and '8'. This depends on your deck (some have a '1/2' card). It doesn't matter, as long as all participants have the same set of values.

Explain the rules: Choose a card for each suggestion. Choose a low value if the action is not worth doing in your opinion. Choose a high value if the action is worth starting next iteration.

Give them a minute to sort out their internal ranking and then present the first suggested action. Everybody chooses a card and they reveal them at the same time. Add the numbers from all cards and write the sum onto the action. Remove the used poker cards. Repeat this for all actions. If you have more actions than poker values the players can show 'no card' (counting 0) for the appropriate number of times.

Implement the action with the highest sum in the next iteration. Add more actions only if there's team consensus to do so.

3 for 1 - Closing: Was everyone heard? (#71)

Check satisfaction with retro results, fair distribution of talk time & mood
Source: Judith Andresen
Prepare a flip chart with a co-ordinate plane on it. The Y-axis is 'Satisfaction with retro result'. The X-axis is 'Equal distribution of talking time' (the more equal, the farther to the right). Ask each participant to mark where their satisfaction and perceived talking time balance intersect - with an emoticon showing their mood (not just a dot). Discuss talking time inequalities (and extreme moods).

(#)


Source:
Retromat contains 127 activities, allowing for 8349005 combinations (25x30x22x22x23+5) and we are constantly adding more.

Created by Corinna Baldauf

Corinna wished for something like Retromat during her Scrummaster years. Eventually she just built it herself in the hope that it would be useful to others, too. Any questions, suggestions or encouragement? You can email her or follow her on Twitter. If you like Retromat you might also like Corinna's blog and her summaries on Wall-Skills.com.

Co-developed by Timon Fiddike

Timon gives Scrum trainings. He mentors advanced scrum masters and advanced product owners. Human, dad, nerd, contact improv & tango dancer. He has used Retromat since 2013 and started to build new features in 2016. You can email him or follow him on Twitter. Photo © Ina Abraham.