Knowledge sharing in large groups of Scrum Masters is a challenge. Prior to the Covid-19 outbreak it was an issue in SimCorp, and the pandemic did not improve this situation at all. All collaboration happened online, and the challenge for SimCorp’s 40 Scrum masters grew. The authors, internal Scrum Masters, and an external Agile Coach, came up with a series of experiments to enable Scrum Masters to learn from and inspire each other across physical borders and functional areas. By combining known techniques from the Liberating Structure Troika Consulting and the Gemba Walk, triads of Scrum Masters would collaborate closely, gather experiences and consult using a lightweight, simple-to-follow framework.
Focusing on three events, Sprint Planning, Daily Scrum and Refinement, the triads visited each other’s events as a Gemba Walk, and subsequently gave constructive feedback using Troika Consulting.
Gemba Troika Consulting enabled Scrum Masters to overcome their reluctance to share positive and negative experiences. What is more, it increased participants’ awareness of unknown unknowns, of different views on similar challenges, and gave opportunities to share and align both success stories and failures in small autonomous groups of peers across the 700-person Product Division.
Back in 2015 SimCorp started a transition from more traditional project management to an agile setup. Using SAFe as a framework, 500+ product development people moved organizationally into smaller units – seven Agile Release Trains (see the experience report “What SAFe doesn’t tell you”[i] for more details).
SimCorp has been on a journey of doing and being more agile for seven years. Tremendous effort has gone into nudging the change of the culture and mindset of the Product Division, consisting today of 800 people and 80 Teams.
For years SimCorp has had many distributed teams spread over most of Europe running virtual Scrum Events so online meetings were common and handled in a professional way using Microsoft Surface hubs in different countries and company locations.
Over the past couple of years, due to Covid-19 outbreaks, every meeting was conducted online. These virtual meetings in Microsoft Teams led to a decline of scrum event transparency for people outside the teams by excluding architects, coaches, managers, subject matter experts and also fellow Scrum Masters.
A new culture emerged. We began to interact less and less on daily basis and due to the virtual distance, we forgot to make use of the collective wisdom in the community by working together on improving known unknowns, finding the blind spots and aligning to help find and remove systemic impediments. You could say we had created new silos of knowledge in the company. Even when Scrum Masters in an ART did have sharing sessions, cross ART sharing was missing and the things that were shared were only success stories, fostering an “us and them” culture.
Figure 1: People must connect to help each other
3. Experiments – What we did?
In an attempt to challenge this culture in SimCorp, we performed an experiment. We did not set out to create a whole series of experiments, but as it turned out, the experiments we did developed organically, with one experiment naturally leading to the next one on a path.
3.1 The first experiment – Gemba walk
This first experiment was conducted in one ART and with those Scrum Masters that were open and willing to try this new thing out: a few Scrum Masters volunteered to have an external observer at the sprint review, the scrum event where they already had external participants providing feedback on the product increment.
The external observer did a Gemba walk. When you Gemba walk, you “go to see where the work is done,” which is a great strategy, in part because it inspires people to learn from each other and motivates them to become better. Non-judgmental observation, rapport and a shared understanding are essential for any future collaboration. It is crucial to focus on “what you see (observe)”, and if you are in doubt, ask the client Scrum Master to help to put more words on your observations.
3.2 The second experiment – a proof of concept
Involving other Scrum Masters in the work of a Scrum Master seemed like a great strategy. A couple of Scrum Masters were inspired by the concept of combining the Liberating Structure called Troika Consulting[ii] and Gemba walk[iii]. Renato Claudino and Taghrid Elashkr started to formalise the concept of Gemba Troika Consulting after having tried it out themselves and discovering the good, the bad and the ugly related to this concept.
Later both presented the experiment in the internal SM CoP and asked for volunteers to form additional Gemba Troika Consulting groups. Even with a compelling presentation of the advantages of the practice, only a few Scrum Masters volunteered to try it out.[iv] Thus, the aspiration of finding a new way to improve the company culture with this practice failed, since few Scrum Masters were willing to take part in this experiment actively.
3.3 The third experiment – Scrum Master Training Program
An opportunity presented itself: a mandatory Scrum Master Training program for all Scrum Masters in SimCorp. Including the practice of Gemba Troika Consulting in the training seemed obvious. The 40 Scrum Masters were divided into 3 groups (cohorts 1, 2 and 3) and each cohort was subdivided by self-selecting into groups of three people (trios). All trios were instructed to Gemba walk at each other’s scrum events of sprint planning, daily scrum and refinement during the next 2-4 sprints. As an output, each Scrum Master had to conduct three scrum events (as the client, with two observers) and participate in three scrum events twice (as an observer).
3.4 The fourth experiment – How to do “Consultancy”
From the first cohort in the third experiment, we learned how important it is to have concise instructions of what “consultancy” means and how it’s done in this practice. When we invited the remaining cohorts of Scrum Masters to participate in the Gemba Troika Consulting we added extra instructions, to help and guide and in the end ensure the consultancy part in the trios was done in as non-judgmental a way as possible. When the two observers offer consulting, it is a conversation or debate between two people, neither of whom know absolute truths, but they share observations, positive ideas, advice and express their curiosity about the future. The observers must refrain from passing judgement, being negative, talking about shortcomings or suggesting detailed solutions. Most of us know about our shortcomings already and do not need to be reminded about them (they are already behind us). What we need is a positive focus on where we can continue to become even better, building on positive psychology and focusing on what is in front of us.
3.5 Results and Outcomes
Gathering data to consolidate results was done mostly through qualitative interviews in feedback sessions, using online survey tools and running UX Fishbowls[v] to create space for a shared understanding of the good, the bad and the ugly. Gemba Troika Consulting has been an overwhelming success – and an eye-opener for most. Talking to and consulting in troikas enables learning from peers. The learning is centered around everyday challenges – not abstract learning.
Many trios have continued to practice Gemba Troika Consulting after the training ended. They have found sharing across functional areas valuable, with insights like the fact that many teams struggle with similar issues due to systemic impediments continuing to emerge.
Members of Gemba Troika Consulting help each other see blind spots or unknown unknowns. The number of unknown unknowns that have been revealed has been enormous, and this outcome was impossible to predict. If you make space and time for creating awareness and free up debate in small, autonomous groups, the need for change becomes apparent. The ability of a group of people to uncover blind spots is unpredictable and often surprising. Rescinding control increases flexibility. The more flexible the organisation and the people within it become, the more likely they are to effect change that is wanted. For this reason, we wish to continue to decentralise and expand our ability to do so.
Figure 2: A simplified illustration of how the observers also learn from the Gemba Troika Consulting
The awareness of knowledge and capabilities of our Scrum Masters are high. As one participant reflected,
“The best part of the training is to be able to join other teams’ events and have others join mine, to get and give consulting, to see how my Scrum Master colleagues work and learn from each other.”
When peers relate to you and your problems, when they offer you consulting instead of direct feedback on what you do (or ought to do), you will listen, gather inspiration, be curious and even try out the suggestions you receive, thus extending your self-awareness. This change is one of the significant findings in these experiments. With it, we have a stronger community, more open to sharing the truly relevant issues and a willingness to explore how we can solve them together. It’s easy to pick up and continue with Gemba Troika Consulting as new issues become apparent.
As an example, one of our shared learnings is that using yesterday’s weather to create a plan for the future within the boundaries of organisational complexity is not optimal. Thinking probabilistically is a better approach to predictability, as it means acknowledging that there is more than one possible future outcome for how something is going to end once you’ve started on that something. There is more than one possible outcome for a particular work item will be complete, given when the work on it has started. If you track your cycle time, you will know what the possible outcomes are for when a work item will be completed, to a certain level of confidence.[vi]
We also experienced an autonomous group of Scrum Masters hosted a CoP learning session to show how it is possible to gather this data and how to use it going forward. It has also created a shared understanding that cycle time and throughput should be a part of our Obeya[vii] (the cockpit monitor for Product Development teams) in the future.
Resistance to Gemba Troika Consulting practice has been minimised. Many Scrum Masters had a clear aversion towards external people joining their scrum events and giving feedback, or even better, consulting. We know how most people feel when somebody asks, “would it be ok if I give you some feedback?” or “I have not had the chance to give you feedback from the last session.” We respond by saying “I’m curious to hear what you want to tell me”, but often our true feelings are closer to “What have I done wrong?”. The consulting part of the Gemba Troika Consulting ensures an open-mindedness towards and awareness of our own competences and capabilities, in a psychologically safe way. This practice has increased the willingness of Scrum Masters to participate in the experiment, try out the practice and do their best to provide their peers with consulting.
If you focus on more than what you observe, you risk missing out on where the magic happens, when the client is open to collaborating on seeing his or her situation from a new perspective. Therefore, try only to finish the statement “What I really like is…”, with focus on being non-judgmental and asking opening questions. This creates psychological safety and makes it possible for the client Scrum Master to see and talk about issues they previously would have found impossible to discuss. They will feel comfortable sharing, increase their shared understanding, and remove biases towards observers.
We experienced a few shortcomings, most of them related to elements already in the system, for instance, how difficult it can be for Scrum Masters to prioritise participation in the other Scrum Masters’ scrum events and find time to do the consulting session over the everyday work of a Scrum Master. As one Scrum Master remarked,
“…the training did not fit into my calendar at all due to end of year holidays. As a result, vacation took a lot of time from the training. Gemba Troika Consulting is a great idea and really helpful, but it’s almost impossible to fit others’ meetings into my own calendar or get everyone in the troika together. “
Loss of control is something you as a trainer must accept to lay the groundwork for psychological safety in trios. We as trainers had no control over the learning or the performance of each trio. Instead, we gathered knowledge of what happened in the form of storytelling from the participating Scrum Masters.
4. How to arrange it for 40 people
A larger group of people had been evolving this practice and running experiments. Most of the people involved are Scrum Masters who have tried out the practice themselves. To drive the experiments, SimCorp engaged with External Agile Coach Jakob Schmidt Sørensen, in total four experiments have been executed so far. The facilitated training (experiments) was run by a group of internal Scrum Masters who volunteered to take an active part. In experiment 3 and 4 the group of Scrum Masters organised themselves in a design team, met regularly and facilitated the activities, workshops, learning sessions and feedback sessions together.
The concept was evolved by Jakob Schmidt Sørensen after having been asked by management to begin consulting the Scrum Teams from a specific ART. This led to the need for a less intrusive way to observe a team’s events as team members and the Product Owner explicitly asked why Jakob was in their meeting not leaving any doubt that his presence was tolerable but annoying.
A smaller group of Scrum Masters tried out a new format initiated by Jakob Schmidt Sørensen. It was based on more peer-to-peer positive feedback. Renato Claudino, Taghrid Elashkr, Stine N. Olesen and Oleksii Tkachuk took part by doing the proof of concept. The learning was that peer-to-peer positive feedback was very useful and had a nice side effect of less resistance in the team for the internal Scrum Masters to take part in their events. For a longer period of time, we tried to get more trios to form but it was difficult for people to prioritise this way of improving as it required effort over and above scrum events, CoPs and other strategic initiatives.
Third and fourth experiments
In the autumn of 2021 management asked for a structured, facilitated training program to help Scrum Masters align and become better at planning and executing. This was a golden opportunity for making a large-scale experiment using Gemba Troika Consulting. All of the Scrum Masters were self-selected into groups of three and instructed to run Gemba Troika Consulting over 2-4 sprints for sprint planning, daily scrum and refinement meetings.
Figure 3: We have added some of the positive feedback from the participating Scrum Masters regarding the Gemba Troika Consulting experience.
The design team was primarily a group of internal Scrum Masters dedicated to creating, designing, and facilitating the learning sessions. Many feedback sessions using Liberating Structures like Shift and Share and Wise Crowds were used along the way and the first training module ended with a UX Fishbowl to gather feedback on the initiative.
5. What We Learned
We have gone from a situation where nobody saw or shared everyday work with teams to a culture where Scrum Masters prioritise Gemba Troika Consulting, i.e., a culture encouraging being curious, helpful and where we strive to share, learn and align.
Scrum Masters now know that their teams are not the only ones experiencing the problems they experience. Often one or more of the other teams have similar issues and like to help or get help finding solutions. This makes setting up solutions experiments together possible and people are willing to share their learning, failures and successes.
The number of unknown unknowns that have been revealed has been enormous, and this result is something that was impossible to predict. By accepting the fact that the design team had no control over the outcome from the Gemba Troika Consulting groups, we set the Scrum Masters free to discover what they need in their given context to gain knowledge, get help and inspiration to move forward.
The autonomous setup fosters an understanding that Scrum Masters can use this practice whenever they like, and that they can plan and execute a Gemba Troika Consulting. There is no need for central coordination, no need for management interference. They can just do it.
6. What’s next
Looking at similar roles in the company in order to try out Gemba Troika Consulting, we have started two troikas amongst the Development Managers. They expressed great enthusiasm for trying it out with meetings like 1:1, interviews of new candidates, strategy meetings, leadership meetings and more. We are waiting with great anticipation for what these two trios will learn from this experiment.
We wonder if this practice, Gemba Troika Consulting, could be used by other professional circles within the company. Others have wondered if it could be used between companies with similar challenges (like small “ERFA” groups) or in other professions, e.g., fitness instructors, carpenters, or sales representatives. Maybe it is a way to create more self-awareness, alignment, and transparency anywhere and everywhere?
We would like to thank SimCorp for daring to run these experiments and pressing on to improve iteratively. Malene Krohn and Piet Syhler have both shown a lot of interest and invested time in the facilitated training and its potential.
Thanks to all members of the design team which also included Pawel Olchowik, Christiane Schabarum, Kristian Adams, Galyna Olikhovska and Olena Levchenko, who have taken an active part in creating, designing, facilitating, and participating in as well as continuously improving the training.
Thank you, Frank Olsen, and Tomas Gustavsson, for supporting us throughout the process.
Finally, a big thanks to Niels Harre, our eternally positive, helpful, and forward-looking shepherd.
 Rapport is a French term translated to something like “the basis of meaningful, close and harmonious relationships between people.”
[i] Article: Experience report “what Safe doesn’t tell you (https://www.agilealliance.org/resources/experience-reports/what-safe-doesnt-tell-you/)
[ii] Liberating Structure – Troika Consulting (https://www.liberatingstructures.com/8-troika-consulting/)
[iii] Article: Uncovering blind spots in your Scrum team (https://medium.com/@renatoclaudino/uncovering-blind-spots-in-your-scrum-team-b8e5b4019e31)
[iv] Article: A Troika Consulting like experiment for daily standups feedback (https://medium.com/@renatoclaudino/a-troika-consulting-like-experiment-for-daily-stand-ups-feedback-5dcad40e4b91)
[v] Liberating Structure – User Experience Fishbowl (https://www.liberatingstructures.com/18-users-experience-fishbowl/)
[vii] Article: What is an Obeya (https://en.wikipedia.org/wiki/Obeya)