In the latest episode of the Agile State of Mind Podcast, I sit down with fellow ProKanban trainer
to discuss the visualization of flow metrics in Kanban. Our conversation reveals practical insights into improving predictability and system health that any software development team can benefit from.We help you make sense of what we see on different metrics charts — cycle time scatterplot, throughput run chart, and estimation vs cycle time!
We discuss how to visualize metrics effectively, the significance of cycle time and throughput, and the role of service level expectations (SLE) in improving predictability. The conversation also highlights the importance of work item age and how it can impact team performance and system health. The episode concludes with a call to action for teams to embrace these metrics to enhance their workflow and decision-making processes.
This is a continuation of the Flow Metric podcast series. In the previous episode, I spoke with Colleen Johnson about why estimations are a waste of time. We unpacked Flow Metrics, how they save us from the never-ending estimation cycle, and where they show up in everyday life. We also explored the stages of grief teams go through when they first see the real numbers behind their workflow.
The Estimation Problem
The episode starts with a striking visualization: a scatter plot comparing story point estimates to actual cycle times. The revelation? There's virtually no correlation between the two. Items estimated at 21 points sometimes finish in just three days, while five-point tickets can drag on for weeks. This inconsistency undermines the very purpose of estimation - predictability.
"We should not try to get better at estimating, but use flow metrics and probabilistic forecasting instead."
Benji explains, capturing the core message of their discussion.
Why Should You Care About Flow Metrics?
If you’ve ever been asked, “When will this feature be done?” and had to rely on gut feeling or best guesses, you already know the pain of traditional forecasting methods.
Flow Metrics provide a data-driven way to answer that question with confidence. They help us understand:
Cycle Time – How long does work take from start to finish?
Throughput – How many items are getting done per unit of time?
WIP (Work in Progress) – How much work is in progress at any given time?
Work Item Age – How long has something been in progress, and is it at risk of taking too long? (Perhaps the most important yet least discussed metric.)
With these insights, the teams can see bottlenecks, adjust priorities, and make forecasts for stakeholders.
From Metrics to Action
What makes this conversation especially valuable is the focus on practical application. Rather than just collecting data, we discuss how these metrics change team conversations:
Instead of asking, "How many points is this?" teams can ask, "Can we complete this within our SLE of 10 days?" This shifts focus from abstract estimation to concrete delivery capabilities.
When aging items approach your SLE threshold, they can be color-coded to signal the need for intervention. This creates natural opportunities for teams to collaborate and prevent items from becoming outliers.
"If Work Item Age is the only thing you talk about in your daily,
that is time well spent."
"There is this joke that, every time you start an item, you put the banana peel on that item. And then you see how it ages and it doesn't look good. It probably doesn't smell good over time either!" Benji explains with a memorable metaphor.
Breaking Free from Estimation
Perhaps the most liberating aspect of this approach is moving away from what Maria calls "the estimation mess." By focusing on flow and actual performance data, teams can make more reliable forecasts without the overhead and false precision of traditional estimation techniques.
"Plans based on average assumptions are wrong on average."
Maria quotes the book Flaw of Averages by Sam Savage, highlighting how conventional approaches often fail.
Watch the Podcast & Level Up Your Kanban Game
This episode is packed with real-world examples, practical insights, and a few laughs along the way. Whether you’re a Product Owner, Agile Coach, Team Lead or just someone who wants to make work flow better, you’ll walk away with actionable knowledge.
If your team is struggling with estimation challenges or seeking more predictability in your delivery process, both Maria and Benji are ProKanban trainers who offer hands-on training to help teams master these principles. Their approach focuses not on selling a "magical solution" but on providing practical techniques to improve your system's health and flow - check out the APK training here!
As we note in the episode, this isn't about achieving perfection but about becoming "least wrong" in an inherently uncertain environment. The journey starts with understanding a few basic principles that can transform how your team works.
💡 Want to learn more? Let’s train your team in Kanban! Contact us for details!
Links to
articles and flow metrics visualization tools:
Share this post