Jennitta Andrea described her background of Anthropology. She was very interested in retrospectives based on her background. Jennitta worked the audience through a simulation. Every paired up with someone else, stand back-to-back after introductions and change five things. Then turn around, and have to find the five things the partner changed. Jennitta explained, that during a retrospective this skill of observation and change help to see improvements.
Jennitta worked some volunteers through a game of jenga, where everyone could take one piece from a tower, and get back to audience after that. She showed two towers of jenga, one pretty finely build, another one looking unstable. The question was, on which project the audience would like to be on. The Purrfect Project would be pretty boring. Still, with a failure to observe, such a project can quickly go down. On a Bad Project out of shape, the situation may seem hopeless, and go down the line very quickly.
Jennitta pointed out to fix one problem at a time, together. This continuous improvement cycle is crucial, not only to retrospectives. She showed the double-loop learning effect, in short iterations, and on a bigger level after release of the product. For little roadblocks the Daily Standup meetings can help to get them out of the way. She referenced material from Norm Kerth – Project Retrospectices, as well as Esther Derby & Diana Larsen – Agile Retrospectives.
Jennitta described that the first part of every retrospective is to get everyone aligned, to see where they stand, and what to get out of the retrospective. When people start to understand individual perspectives, they quickly find out how to work together properly. Building synergy and helping the team see the big picture is crucial to a retrospective.
Jennitta continued to summarize the two mentioned books. A retrospective consists of the five phases which she calls Safety, Discover, Analyze, Plan and Close. How much will each person participate is part of the Safety phase. In general it should not take a very long time to gain that safety. In this you want to have everyone feeling safe to say his opinion, and don’t have a gap in your picture of the last iteration or release. Part of safety is built into the agenda of the retrospective. The opening starts with appreciations. As a homework, think of an appreciation for everyone on the team. She showed how to use a ball of yarn to create connections between the participants – emotionally and physically. She referred to introverted people and extroverted people. Extroverts need to talk in order to think, while introverts have to think before they talk. A good retrospectives supports both modes of thinking. Jennitta pointed also out that early participation in an retrospective leads to continuous participation. So, everyone should speak up in the first five minutes to create safety to speak with this.
Jennitta pointed out that saftey is built into the ground rules of retrospectives. She referred to the prime directive from Kerth:
Regardless of what we discover, we understand and truly believe that everyone did the best job they could, given what they knew at the time, their skills and abilities, the resources available, and the situation at hand.
Jennitta uses safety builts into team norms. She uses index cards filling up the fragment “I don’t like when…” Then they exchange cards with other participants, who then come up with suggestions on the problem stated. Thereby team norms are created collaboratively in a helpful way. She said this is a simple technique that provides safety. She showed Kerth’s anonymous poll technique to get the feedback from participants on how safe they will. The facilitator then gets a picture on how safe everyone feels to speak up, and how comfortable they are to say what they think.
Jennitta described the second phase of the retrospective, where the team discovers what had happened during the last iteration. She explained that people should build affinity groups for this activity. She presented the timeline activity from Agile Retrospectives for the discover phase. She explained how to set up the timeline, to put on past time periods, have team members put on sticky cards on the timeline, showing whether they were happy or sad with something, as well as an emotional graph on the bottom. The different variations on this timeline exercise are infinite, so you can play around how you design it. It should get you clarity though, in order to get the big picture. On how to build such a timeline, she explained that you can do it by building all at once, or time segment by time segment, or do the emotional seismograph first.
As an alternative exercise she showed fingerprint graphs as a team radar. She showed a focused simulation. Describe what happened, account for your feelings and actions, transfer this to the real world, and action plan for the changes you want to bring in. This approach helps to collect the data you want to see in order to improve later. The simulations help to see things we haven’t seen before.
On the analyze phase, Jennitta showed the points continue, stop and start. What went well, what went bad, and what to try out. A balanced approach is to see what the problems are fist, define the problem, fix what’s broken, and focus on decay. She explained that you want to review the timeline if you built one. There are different variations on how to do this. If you built it all at once, then you might want to review it segment by segment. The same holds if you built it segment by segment. If you did the seismograph first, then you might want to review this one first to see patterns.
On the plan phase Jennitta explained that we want to find out where do we want to be, and how do we get to be there from where we are right now. One thing teams have to keep in mind is that we cannot everything at once. Each person therefore gets five dots for dot voting of what the priorities for problem areas are. Prioritization affects priority, Jennitta explained. A facilitator must explicitly remind about the group goal to get alignment. Once we have a prioritized list we spend a short amount of time on identifying what to improve. She showed a technique to use a flipchart sheet, dividing it into four quadrants with “Where are we now?”, “Where do we want to be?”, “What hinders us?”, “How do we get there?”. The outcome of this should be concrete action items your team is going to work on during the next iterations.
For the closing Jennitta recommends to get in a circle so everyone can see each other, review the action items, and end the session well. At the end of her keynote, Jennitta explained where to focus on dependent on the kind of retrospective you do. For iteration retrospectives you may want to focus on team & process issues, for daily ones on removing roadblocks for your team. Also for daily and iteration retrospectives you invite the core team as participants, while you bring in contributors across the organization on a release retrospective.