Fire, Water and Retrospectives

Osama Al-Eryani
SSENSE-TECH
Published in
8 min readMar 21, 2019

What I learned from starting and facilitating retrospectives for 9 teams.

A simple retrospective with a clear grouping and an action item proposed by a new team member

Retrospectives, or simply retros, are structured conversations in which a team inspects and adapts how they work together.

My context is working in Tech teams, but retrospectives can just as well be used for HR, Accounting, Sales, Marketing, C-Level teams, or even in a family setting.

Background

I used to work in a large team as an individual contributor. It was less of a team and more of a group of individuals, each working on their respective tasks, with a coordinator managing the integration of their work. Plans were drafted, with hourly precision for the different phases.

Thanks to the notorious unpredictability of our line of work, it was common to be behind schedule by months. By the end of the project — say a year later — we would all sit together to discuss how things went.

A special retro for a new team, with framing questions designed to anticipate further problems down the line

Reflecting upon this, I figured there must be a better way to reduce, or at least acknowledge, this uncertainty.

I proposed that we split our group into three smaller teams and gradually shift from monolithic releases every six to nine months, to monthly, and finally to bi-weekly. This required adjustments at all levels: our stakeholders realized they did not need to squeeze in all their requests before we started; rather, they could wait for the next cycle. The group members realized they were now a team. They knew each other’s names, what everyone worked on and how they could support one another. Lastly, our managers gradually let go of chasing specific dates, in favour of a more constant stream of value addition.

In this dynamic setup, we found ourselves adjusting how best to work together, which did not always go smoothly. The managers set the overall constraints but we still needed a lightweight mechanism for resolving disputes and continuously evolving as a team.

Enter Retrospectives.

Retrospectives = Process Kaizen + Team Therapy

Praise section makes an appearance

Here is what I learned from facilitating retrospectives for nine teams over a three year period:

1. Get Management’s Blessings

Introduce the concept of retrospectives to the team’s managers.

Retrospectives are a mechanism to increase the team’s autonomy. It does that by surfacing and addressing disputes — before they escalate — revisiting working arrangements as needed, reinforcing the good stuff, and continuously improving how the team works.

Note to my early self: managers are not the enemy. They provide care and support that nurtures the team, help with what is out of their control, and bring in strategic work that keeps the team engaged.

Note to my current self: be wary of too much autonomy in teams with limited technical expertise or an unclear business vision.

2. Find Time in the Team’s Schedule

I have observed that teams cancel retros when they need them the most. Here are a few ways of mitigating this problem:

  • Mandate it as part of your engagement as a facilitator. Note that, depending on your environment, this might be a bit too coercive.
  • Embed a small slice into existing meetings, show value, then expand. This works almost everywhere, but with a slower ramp up.
  • Channel the team’s process changes and points of contention to the retro forum.
A timer proves super handy for speedy retros

3. Find a Facilitator

If you are a manager interested in starting retrospectives in your team, it helps to pick a facilitator from outside the team who is detached from the specific outcomes of the discussion.

We tried having team members facilitate their own team’s retro, but they often found themselves torn between the requisite neutrality of facilitating the discussion and their desire, as a contributor, to push for certain outcomes.

We have since experimented with having team members facilitate retros for other teams, and the early results are promising.

The facilitator’s role:

  • Invites as many different views as possible into the discussion
  • Ensures procedural fairness during the discussion
  • Helps the team converge towards specific and previously undetermined outcomes

4. Design a Framing Question

Freshening up the framing question

The purpose of the framing question is to gather facts and sentiments regarding different events that transpired during the period in question.

Examples:

What went well? What didn’t go so well? Ideas for Improvement

Good Not So GoodIdeas

StartStop Continue

😊 😣 💡

👍 👎 🌷 💡

Who are we as a team? What is the objective of the team?

Great Not So GreatThank You’s Help Requests

A good facilitator will pay special attention to how the team speaks, and choose language attuned to the team’s culture.

Asking “What was bad?” versus “What was not so good?” could be the difference between a lively discussion and silence.

New grouping “Help Requests” emerged during the retro

Lastly, avoid defining your question too narrowly. Doing so could exclude topics wherein the real problems reside.

5. Conduct the Retro

Reflection

Ask your framing question to the group, hand out post-its, and encourage silence. Observe body language cues, this can help anticipate what is to come. If you have new team members, take advantage of their fresh perspective and encourage them to write.

Presentation

Once everybody has written on their post-its, announce “the floor is now open, for whoever would like to start”.

A retro examining a specific incident which, after gathering all the information and generating multiple hypotheses, led to an unexpected solution

At this point, the team members will take turns to stick their post-its to the wall, read them aloud and explain their significance. Be wary of people sticking up their post-its all at once before explaining them, this makes it harder to refer to them later for further treatment or prioritization. Another thing to watch out for is back and forth discussions, other than clarifying questions, before everybody has presented. This is the fastest way to derail a retro with potentially less important topics. The team’s energy is finite, save it for the most important topics.

If an emotional presenter looks at you, nod and acknowledge, but avoid engaging just yet. Once everybody has presented, you will have the full picture including everyone’s feelings, observations, and suggestions.

This is your raw material.

Grouping

As a facilitator, having listened to the group, and read the post-its; try to identify clear patterns. Are there different post-its branching from the same theme? If yes, ask the group to confirm. If they agree, group the post-its by themes and draw labelled circles around each group.

The theme labels can influence the forthcoming discussion, so choose them carefully, and remain open to changing them. You could also ask the team to help you name the themes.

While not always necessary, you could create themes for “good things” as well. Especially if it is recognizing someone!

Once the biggest themes have been identified, you could ask the group to come closer to the board, and do the same for the remaining post-its. By engaging the team in this way, it does three things for them: firstly, they practice creating order out of chaos; secondly, they learn to deprioritize that which doesn’t align with any of the main themes; lastly, it gets them warmed up for the next stage — discussing solutions for the themes identified.

Creating order out of chaos

At the end of this phase, your board should be shaping up, and themes about the major pain points should be emerging.

If you were to stop right here, it would already be an achievement. You have a holistic view of the problems, with an energized team, invested in solving them. This level of understanding may not have existed before.

A complicated retro, with the same theme generating opposing sentiments (notice the lack of written action items)

Spark the Fire 🔥

You have the haystack (the themes) and the matches (the people), all that is missing is a little controlled fire.

Sometimes it is helpful to start with an uncontroversial and large theme first to get the team comfortable with the discussion.

For this, you could select the largest theme and ask your team what they think can be done to resolve the problem at hand. If this is followed by silence, remain patient. The longer the silence, the deeper the thoughts.

Delay the attack until the team is intimate with the problem.

Eventually, somebody will make a proposal. If it is directed at you, subtly redirect the question to the team. Give the team a moment to react. If nobody does, pick someone at random, and provoke a response.

Mixing elements from different retros, for a squad reboot

It is unlikely that the second person will agree with no further additions. They will likely add their own insights, which will sharpen the initial proposal, or create stronger alternatives.

Once the second person has spoken, usually the rhythm of the exchanges will pick up pace.

The fire is now lit.

The goal of the fire is to spark multiple points of view, and forge them in the heat of conversation.

Your job is to keep the fire burning at the right intensity. If it is too weak, nothing of significance will be discussed; if it is too strong, team members might become wary of opening up in the future.

This exercise allows team members to openly discuss volatile matters that might otherwise lead to more toxic outcomes. It also makes them more likely to adhere to resolutions that were reached by consensus. Lastly, if the fire does not light for a particular topic, give it a rest. Maybe it is not that important after all. Or maybe the team is not ready to discuss it yet.

Action Items — Quell the Fire 🌊

Clear links leading to action items, and named owners

If you channel the conversation well, there will be an idea for each theme that gathers the most momentum. As time runs out, grab onto that idea and ask the team what they think is the smallest step they can take towards it. Write it on the board, and confirm with the team if they are willing to commit to it. For those who remain hesitant, reassure them that the action item and its impact will be revisited in the next retro. Once everyone is on board, your job is done.

Conclusion

The trick to a successful retrospective is to delay judgement until a holistic picture emerges (via the reflection and presentation sessions), and the team collectively deduces the themes in this picture (via the groupings). Then, and only then, should you ‘spark the fire’ to invite judgement, analysis, and finally, plans of action.

This ends the retro with a sense of continuity in the team’s experience, a holistic understanding of the major challenges, a collectively agreed upon plan of action, and the promise of accountability in the next retro.

And that is the magic of retrospectives.

Editorial reviews by Deanna Chow, Liela Touré & Prateek Sanyal

Want to work with us? Click here to see all open positions at SSENSE!

--

--

Osama Al-Eryani
SSENSE-TECH

Transforming Conversations - Agile | Product | OD