Retrospectives: Looking Back to Move Forward

Authored by Amy Lane and Marta Maio

Even if we don’t realise it, we’re constantly inspecting and adapting on a daily basis. You’ve possibly tried cooking a new fancy dish before and it didn’t quite go as you hoped it would — a proper Kitchen Nightmare in real life. Mostly likely, once the sorrow of the failure is over, you’ll be ready to give it another go. For that, you need to analyse what didn’t go so well (too much salt, not enough cooking time) and what you have nailed that you want to keep (great choice of cheese!). We follow thought processes like this so much that most times we won’t even realise it. If that’s the case for our day-to-day, why shouldn’t it be for our work too?

It’s common practice that Agile teams do these retrospectives often, and it makes sense. Why would you want your dish to come out wrong again and again? The cost of not inspecting and adapting may not be glaringly obvious at first, but cracks will soon start to show.

Nevertheless, as with almost everything in life, it is always best if we can look at some data to prove our point. Here are some interesting stats related to the positive impact of retrospectives in Agile teams, from Parabol :

• 81% of Scrum teams report holding a retrospective after every sprint.
• Teams that have regular sprint retrospectives have 24% more responsiveness and 42% higher quality with less variability than teams with infrequent or no retrospectives.
• Teams that conduct effective retrospectives have a 20% higher balanced performance than teams that don’t conduct retrospectives.

Additionally, as a bonus, some fun trivia from Echometer on what a typical retrospective looks like. Turns out that the average retrospective:

• Takes place on a Wednesday, however Friday is the favourite day of the week for teams.
• Lasts around 1h and 11 minutes, however the ideal time will be 30–45 minutes.
• Most times, is attended by 10 people or less.
• Its occurrence is bi-weekly or less.

When you were reading the above, you were probably thinking about the standard way of doing retrospectives: done at the end of the cycle, covering what went well and what didn’t go well and coming up with actions.

Still, retrospectives can take on various formats and purposes beyond the standard format, enabling teams to inspect and adapt in numerous scenarios. Let’s explore some specific examples of how retrospectives can be applied in different contexts to drive continuous improvement:

• Post-mortem/ Washup: an “After action review” — this could be at the end of an outcome as a learning activity for the whole team and stakeholders together i.e. were there specific ways of working that were successful or not, or perhaps after an incident to allow time for reflection, root cause identified and preventative actions to be recorded.
• Health Check: higher-level team health checks can be useful to get a feel of areas of improvement, ranging from codebase quality to teamwork & collaboration. Discussing pros and cons of each area and identifying a plan of action can be beneficial for the team’s day-to-day.
• Retros with stakeholders or third parties: this can be beneficial to have between tech teams and business teams to really make sure everyone understands ways of working and allow third parties, tech teams or business teams to surface challenges i.e. how work comes into teams, challenges on shared processes and co-design any improvements together.
• Ways of Working: can be used to review current ways of working and deciding what to discard, keep or bring in. This should be done every so often as the relevancy of some past decisions get out of date with project, team and company changes quite quickly.
• Team Metrics: inspecting, adapting and taking actions out of analysing team metrics together, especially when there are outliers that the team identifies.
• Retro of Retros: not a meeting about a meeting! Have you ever reflected on how you or others run retros? It can be easy to get stuck in a rut or follow a similar format because it once worked. But if the team has changed, maybe the team has matured and needs different challenges or approaches. Optimising how you retro is a great way to ensure you maximise what you get out.

As you can see, the concept of retrospectives is not one-size-fits-all; it evolves as the team matures. Depending on where a team is on Tuckman’s team maturity scale — forming, storming, norming, or performing — the approach to retrospectives should adapt to meet the team’s unique needs and challenges.

Let’s look at the 4 stages of Tuckman’s model and how retros could be adapted to suit each team stage:

• Forming: it is really important that retros are frequent and have measurable actions. They might want to focus on team building, some basic foundational ways of working reviews and experimenting with different frameworks even to find what suits them best. Confidence building is also key for new teams. It is likely they won’t know each other very well and it is important that the team can build trust between themselves. For the first few retros, you might want to spend more time on icebreaker activities to help build a safe space, and don’t forget to leverage the “Prime Directive”:

“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”.

By getting the team to sign up to this, it can support psychological safety.

• Storming: This phase the team will likely need to be focusing on accelerating continuous improvement. The retros might also want to focus on celebrating successes and regularly look at Kudos to help build momentum and motivation in the team. One idea would be to do some focused themed retrospectives that start to surface insights on data points they can measure between cadences and encouraging experimentation from these insights to help try new things out. e.g. perhaps the team is struggling with aligning on estimates. You might want to review your tickets and their sizes or estimates (story points, T-shirt sizing if relevant) to see if the team would still estimate them in the same way now. Possibly looking at “The story of a story” if there are particular tickets that got blocked or were problematic, and look at all the stages the tickets went through to identify bottlenecks or challenges.
• Norming: As teams move into norming phase, it is expected that the group is more self-sufficient. This could be a good time to challenge the team to take the reins of retrospective facilitation, which should bring more variety and fresh ideas to the rest of the group. It should be possible to start digging a bit deeper into how team performance and metrics play a role in the subsequent topics that are raised. Charts like CFD (cumulative flow diagram) and Cycle Time Analysis can truly help improve predictability and complement the introspection exercise.
• Performing: Building on the behaviours and skills of the norming team, it would be expected that the performing team would be more pro-active with ideas for experimentations and driving actions. It is more likely that the actions are more distributed across the team (everyone has a part to play). The insights could be more in depth and there might be a particular focus on how the team can not only make change within the team, but also how they can influence larger scale change outside the team. It is worth monitoring engagement in the team at this stage to ensure it still happens on the agreed cadence as there may be apathy towards events that they have familiarity with.

Successful retros are the foundation of healthy and happy teams, no matter what level of maturity they are in.

In our experience, these are the things that can contribute to successful retros and, in turn, healthy happy teams at any phase of maturity:

• Variety: Most teams will not enjoy the same thing week in week out. But it is also important to adapt to the team’s needs and environment, otherwise the value of retros will be diminished. Maybe consider a retro of retros to understand what the team wants to achieve.
• Timebox: in order to keep the team engaged and the conversation relevant, respecting the duration of the meeting is important. If you want to keep the conversation focused, keep in mind that usually more negative topics tend to take longer. Allow for flexibility on this by balancing how deep you want to allow the group to go. It may be that conversations get too technical or too specific. In these situations, it is ideal to zoom out a bit and re-focus the group.
• Relevancy: Actions should be the whole team’s responsibility to check in on. Therefore, they need to be documented somewhere, accessible and accountability agreed so the team feel they are getting value out of the actions and driving the change. Having a long list of actions is also not desirable — quality over quantity is definitely the goal. Use SMART actions (Specific, Measurable, Achievable, Realistic and Timely).
• Review: try to avoid having all your actions sitting on the team’s wiki. Review past actions every cycle, or even once a week. By doing this, you make sure ownership is present and not forgotten, and we are taking them seriously. To make it fun, why not compile an end of the year summary and gather some metrics, such as number of actions done from retros?
• Rotate: keep things interesting and different. Why not rotate facilitation each time? It is not only a great opportunity for people that normally don’t get to run meetings to do so, but it also keeps it fresh for the rest of the team. It is a way of sharing responsibility and ownership, and to have diverse retrospectives every time.

In a team retrospective, using powerful questions is a great tool for encouraging deep reflection, promoting open dialogue, and driving meaningful improvements. These questions encourage team members to think critically about their experiences, uncover underlying issues, and explore new perspectives. Powerful questions usually start with “Who, What, When Where or Why”, and by nature are open ended. This provides a platform for individual insight, not biased by an opinion of a facilitator to elicit feedback on what worked well, what challenges were faced, and how processes can be refined to be better next time.

Examples of powerful questions, outside the normal 3 retro questions:

• Who haven’t we collaborated with recently?
• What could be a different perspective/angle on this?
• If you could do anything you wanted, what would you do?
• What else? (this is the golden coaching question to elicit deeper thinking)
• What is the opportunity here?

These questions can also be useful in helping people to design solutions or flipping a perspective perhaps turning a feeling of negativity into discussing what opportunities for change there can be. A question like “what does good look like?” can transform a conversation from focusing on what went wrong to handing ownership to suggest something better.

In conclusion, retrospectives serve as a cornerstone for continuous improvement, enabling teams to reflect on their experiences, learn from their challenges, and celebrate their successes. Whether a team is just forming or well into the performing stage, the flexibility and adaptability of retrospectives ensure they remain relevant and impactful. By varying the format, timeboxing discussions, and focusing on meaningful actions, teams can maintain engagement and drive meaningful change. Ultimately, retrospectives are not just about looking back but about moving forward with intention, fostering a culture of collaboration, accountability, and continuous growth.