Over the past few months I worked with a local startup as their Scrum Master. The CEO wanted to make his company more agile — just him, his daughter, another employee, and a few student helpers.
But what does “agile” even mean? In our first video call, I walked him from the old rigid world where plans get executed stoically and deadlines met religiously, only to end up late, over budget, and building something nobody wants. Into an agile world where you produce what stakeholders actually need. Something with real value.
That naturally has massive consequences for how you collaborate and prioritise work.
But that’s not what I want to talk about today. I’d rather capture my lessons from this project. Because there were several insights. The most important one (and honestly not that surprising): habits and people don’t change overnight.
Neither do I.
1. Take more time for deeper briefing: After our first conversation, we jumped straight in. I found myself in a video meeting with the team and the CEO immediately handed me control: “Right, Reiner, tell us about agile and Scrum.” All alarm bells should have gone off. Obviously the team knew nothing about me, so I had to “sell” my ideas and myself first.
A particularly thankless start. Because as a good Scrum Master (I know this now), it’s primarily about listening, not showboating.
I should have cancelled the meeting and aligned with the CEO first. What did he expect from me? Where was his team at? What did he want to achieve? Did he have a purpose, a roadmap, any idea what he wanted to offer and to whom?
Turns out: none of that existed. If I get thrown into the arena unprepared again, I’ll start with general questions or a brief Liberating Structure. Definitely not talking so much about myself.
2. Smaller steps toward the goal: Startups want to move fast, “muddle through,” improvise kamikaze-style. Isn’t this absolute flexibility that follows the boss’s mood also part of agile thinking?
In my interpretation, truly agile people have a backbone. They don’t act like flags in the wind. That only works when it’s transparent enough what you’re doing and why, and what you’re better off not doing because it won’t get you there.
For that, the entire team needs clarity on why and what they exist for, which principles they follow, and who they’re producing something for. Purpose, some collaboration rules, and stakeholders, stakeholders, stakeholders.
I’ve seen too many teams produce something that has no value for stakeholders in that form — or would have had more value if the team had engaged deeply and empathetically with the stakeholders’ situation from the start. And, unthinkably, actually talked to or even involved the stakeholders early on.
How are you supposed to prioritise work if that’s not clear? So I turned up to the second meeting with a stakeholder map. We collected stakeholders and ordered them into an influence/stake matrix. The team struggled to place stakeholders in the most important quadrants. A clarifying conversation didn’t help. Because in the end, the CEO’s voice counted five times as much.
As Scrum Master, you quickly find yourself in a difficult position when you observe that team members don’t have much say. It’s an uncomfortable antipattern. I shared this observation with the CEO in a one-on-one, but fell on deaf ears. He was always excited about the agile idea, but didn’t really want to live it. And several times he ended conversations mid-discussion when the topic got too sensitive.
3. Don’t promise magic tricks: At the start we agreed I’d be there as Scrum Master until year’s end. What can you achieve in three months? We can identify stakeholders, start a daily scrum, maybe set up a backlog board in Jira.
But if two of the three employees won’t make friends with JIRA, then the agile bootcamp gets difficult.
I remember a meeting where the CEO dialled in from his car on the motorway, wanting to populate and sort the backlog with us. A fruitless exercise when it’s unclear which stakeholder the task is meant for (we only got to “user stories” late in the piece).
My lesson: I’m considerably more conservative when it comes to progress forecasts. Better to under-promise and explain that all this takes time, that you should work on mutual trust first rather than turning everything upside down immediately. Watch out for impatient clients. That tends to go badly.
4. One hat is enough: I was supposed to make the company more agile and initially thought of Scrum. In the Scrum framework there’s a Scrum team consisting of a Product Owner, a Scrum Master, and Developers. How does that work when the team (including me) only has four people?
I quickly noticed that not the CEO, but one of the two employees was a born Product Owner. Sharp, quick, agile in the head. But the role was already filled. The CEO was naturally the Product Owner. Yet he didn’t take on the tasks of a Scrum Product Owner: optimising value for stakeholders, knowing the market precisely, prioritising the backlog.
Initially I tried neutrality. I was Switzerland personified, held back as facilitator. But increasingly I wanted to voice my opinion. I signalled this with a cowboy hat on my head. When I wore it, I expressed an opinion, an assessment. I thought I was moving the group forward, but ultimately I disturbed the dynamic by probably interfering too much.
For the next project I want to lead more diplomatically and really only give an opinion when I’m asked twice in a row. At least I now understand that in future I only want to be Scrum Master, not a combination of Product Owner and Developer.
5. Stick with it: For me the most important insight is that I really enjoy working as a Scrum Master and that I can contribute to the team’s success. I keep thinking of Momo the cat, who is just there and everyone confides in her, but who doesn’t actually say anything.
As a result, I’ve refined my own purpose and mission statement as Scrum Master. Another insight: I’d hidden too much behind tools, but really it’s about presence and opening a space where people can think and work together openly and mindfully.
That’s my goal, but I’m still far from it. I’m curious where I’ll get to collect new insights next.
From the archives of reinergaertner.de, running since 1997. Translated with AI help and my questionable bilingual proofreading. If you spot a Germanismus — that’s a feature, not a bug.