BLOG
In 2022, Untis had a problem most growing organizations would recognize.
Too much in progress. Too little shared visibility. Teams working hard in parallel, toward goals that didn’t always connect — and no reliable data on how long anything actually took.
When they counted up the active initiatives across their product organisation, the number was 74. Another 63 were already planned. 59 more were waiting behind those.
Nobody had designed it that way. It had accumulated, the way these things do.
At Agile Tour Vienna 2025, Christian Gusenleitner and Christian Eis — both coaches at Untis — shared what they did next. Four steps, applied over roughly two years, that brought the active initiative count from 74 to 26 and cut defect resolution time from 836 days to 15. Not by introducing a new framework. By following the work.
This article traces those four steps, the data behind them, and what made the change stick.
Both Christians are deliberate about what they’re not discussing. Culture, leadership, self-organization, the shift toward outcomes – all important, all out of scope. The focus is specifically on how they transformed the management of work.
The 2022 situation: autonomous teams with unclear processes and insufficient alignment, very old backlogs, no real data on how long things took or how much capacity existed, and constant friction from trying to squeeze work into two-week Sprints that didn’t fit. Scrum had been tried. It hadn’t landed well – not, they argue, because Scrum is wrong, but because the teams were trying to sprint toward goals they couldn’t see clearly.
The first step wasn’t a transformation. It was a mapping exercise.
Every team got a clear Kanban flow with explicit start and end points. Work in progress was made visible. The key metrics were introduced: cycle time (how long items take to move through the process), work age (how long items have been active), throughput (how much gets done per period), and WIP count.
Critically, they didn’t design new processes. They mapped what already existed – making it visible, measurable, and discussable. The starting point is always the current reality, however uncomfortable it is.
Little’s Law became a working principle: in a stable system, the average number of items in process equals the average arrival rate multiplied by the average time an item spends in the system. In practice, this means WIP and cycle time are not independent variables. Reducing WIP reduces cycle time – often dramatically, without any other change.
Once daily work was visible at the team level, the next question was obvious: what does all this activity actually add up to?
Each team got a second board – a goal-level board showing the larger chunks of work (roughly equivalent to Epics) that daily tasks feed into. Same principles: flow, WIP limits, aging. The framing was deliberately non-prescriptive. Whether teams called them Epics, Objectives, or something else wasn’t the point. The point was having the goals visible as work items, subject to the same discipline as the tasks below them.
This is where the 74 number became visible.
All the team-level goal boards were combined into a single cross-team view. For the first time, everyone in the organization could see simultaneously what every team was working toward. The number of active initiatives was counted. The result was surprising even to the people running the exercise – stickies overlapping because there wasn’t room for all of them.
74 items in progress. 63 planned. 59 waiting. And almost no shared sense, before that moment, of how deep the pile had grown.
A weekly 30-minute alignment meeting was introduced – cross-team, using this combined board as the discussion surface. Not a status meeting. A coordination mechanism for managing the flow of work across teams. Flight Levels thinking, as developed by Klaus Leopold, provided the conceptual framework: different levels of the organization operate at different tempos, but they need to be connected.
The number dropped from 74 to 26 over the following one to two years.
The fourth step addressed something the previous three couldn’t: the gap between what the organization said it wanted to achieve and what the teams were actually working on.
A strategic-level board was introduced, using the same principles. This made divergence visible. Work happening at team level that didn’t connect to any strategic priority. Strategic priorities that had no active work behind them. The misalignment wasn’t a mystery once it could be seen.
Teams pull work autonomously at the daily level. Goal-level alignment happens in the weekly meeting. Strategic allocation is managed separately, by a smaller group, less frequently. The result is what the speakers call a „fractal system“ – the same flow principles operating at different speeds across different levels of the organization. What Actionable Agile calls the difference between team-level and portfolio-level flow.
The defect resolution story is the most concrete outcome in the talk.
Before Untis introduced a „zero open backlog“ policy for defects, the 85th percentile resolution time was 836 days. After – with the same teams, the same tools (Jira and Actionable Agile) – it dropped to 15 days.
Not a different team. Not a rewrite. A different relationship with the work, made possible by visibility and WIP discipline. Practitioners who want to develop this kind of data-driven coaching capacity – understanding flow metrics, facilitating the honest conversations these boards surface – will find TechTalk’s Advanced Certified Scrum Master program covers the measurement and organizational coaching skills that make this kind of change stick.
The cycle time scatter plot makes the before/after visible in a single chart: each dot is a completed work item, positioned by when it finished and how long it took. Pre-change – dots scattered widely, the moving average swinging unpredictably, outliers reaching months or years. Post-change – a tight cluster, a stable average, very few outliers. What changed was WIP limitation and stronger focus.
That’s it. Nothing else.
Both Christians are honest about the hard parts.
Maintaining focus is continuously difficult. There’s always one more initiative that seems too important to defer. WIP discipline requires saying no to things that feel urgent – and that requires organizational will that’s hard to sustain. TechTalk’s OKR Training is worth mentioning in this context: getting the vertical alignment right between what teams are working on and what the organization is trying to achieve turns out to be where most of the gains live, and where most of the discipline is actually tested.
The system perspective doesn’t come naturally. An individual who can’t continue a task and starts a new one is being entirely rational from their own perspective. From the system’s perspective, they’ve just added to the WIP count and made everything slower. That gap – between personal logic and system logic – is a cultural shift that takes longer than any of the tool changes.
And patience. One to two years for the core changes to feel normal. Longer for the data to produce reliable forecasts. You can’t shortcut this with a better framework. The Christians say this plainly, and it’s probably the most important thing in the talk.
The result – 26 active initiatives instead of 74, defect resolution time of 15 days instead of 836 – didn’t come from a new framework. The system at the end has a name, Flight Levels, but that’s not how they found it. They found it by following the work, step by step, from where they actually were.
Christian Gusenleitner and Christian Eis presented this talk at Agile Tour Vienna 2025. The 2026 edition is live – the full program is available now.
Build your solid foundation through our training and continue developing your leadership and coaching journey.