Predictability Is My North Star

0:00A recent student asked me to review an article called Predictability is My North Star. Velocity said healthy, the system said unhealthy. I chose to trust a system. For six straight sprints the velocity chart looked great. Every sprint team hit the number. Every review, Dashboard Show Green. And yet, the last couple of sprints felt bad. Let's stop right there. This is exactly why I don't like using velocity as a metric of health. Velocity just has to do with how much crap you're cranking out. And I think that if you don't understand velocities, but intended to be a measurement of volume, you're going to be sorely mistaken when it comes time to do a health check or a wealth wellness check on your product service.

1:01So, work thrashed, priority shifted, unplanned items kept sneaking in. The team was exhausted, and I needed a way to explain why the system felt chaotic when the metric insisted everything was fine. This is exactly why I don't use the metric. That was the moment I realized velocity was not telling me the truth. I needed a better way to understand what was actually happening. I needed a way to see what went wrong in account for the change. And that's when I decided predictability would be my north star. Yeah, we'll talk a little bit about predictability because that's exactly what I do with shirts and t-shirt sizes. We'll talk about it in a minute. Velocity gave the illusion that everything was healthy. It said we were consistent, it said we were reliable, and it said nothing was wrong. First and foremost, fear of velocities 2020-2020-2020. That's like a death sentence. Dude, feel good up to one of those monitors, right? You should have some ebb and flow. So if it's consistent, and if it's reliable, and if it's always there, that doesn't necessarily mean it's being predictable.

2:02Velocity is a measurement of output, not a measurement of conditions. It can stay stable, it can stay stable, even when everything around it seems to be falling apart. Velocity can be perfect, while the system is burning to the ground. The real questions I needed to answer, I didn't need to know how many points were completed. I needed to know what changed since the last time we met. Why did it change? What happened in the ecosystem as a result that I changed? And can we account for this change next time in some way? Velocity couldn't answer any of those questions. Predictability could. Predictability, in other words, became the answer. I needed a way to explain why things felt bad. I needed a way to show when Sprint Health wasn't as good as it looked. Predictability gave me something velocity never could. Predictability gave me completed plan work divided by plan door. A Sprint with 94% predictability, 29 out of 31, tells a very different story than a Sprint with 62 points completed.

3:04Predictability measures reliability and not speed. It answers one question. Did we deliver what we said we would and if not, why not? More importantly, it reveals the system's behavior. It showed when work added mid-sprint. It showed when priority shifted. It showed when the system was unstable. Even when velocity looked perfect. Predictability exposed to truth that velocity was quietly hiding. You can't do it alone, though. You need to have a leadership shift. I stopped assuming good velocity equals a good Sprint. I stopped letting single numbers define Team Health. Instead, I started looking at why the Sprint felt bad even when the metric looked good. I began measuring predictability or commitment reliability, unplanned work or system interruptions, and flow stability or throughput consistency. I'm looking at limiting width pretty much. These are things that velocity hides. This is where I grew as a systems leader. The payoff, when I shifted to predictability,

4:06I finally had a way to explain chaos. I could show stakeholders exactly what changed and why. I could protect the team from invisible churn, and I could see the system clearly and not just the output. And when teams started looking at predictability, they stopped panning their estimates, stakeholders stopped asking for just one more thing. Roadmaps became believable. Retrospectives became more about learning and not about bullying. Predictability didn't make us faster. It made us trustworthy. And that, my friends, is what changes everything. That's going to do it. I hope that you enjoyed this review. Overall, Michael, congratulations to great article. I'll drop a link in the comments. If you have a topic you want us to discuss, discover or review, please let us know. Learn more at agile dad.com. I'd love to you for me. And as always, I encourage you to stay healthy. Stay well and stay agile. My friends, until next time. Do take care. Bye now.

5:10Bye.