In my 3.5 years of working with diverse teams across various industries, one thing is crystal clear—everyone hopes to leverage agile methodologies to enhance work delivery. With a plethora of options available, it’s puzzling why issues persist. Could it be that everyone is doing it wrong? Or could bad habits be holding them back from success?
I’m betting on the latter, and I’d wager that you’re making at least one of these five mistakes:
1. A Poorly Designed Workflow
A workflow, as the name implies, should illustrate how work progresses through your system. While this may sound straightforward, it can be complex depending on your team’s dynamics. There's no one-size-fits-all formula for creating the perfect workflow, but keeping a few key principles in mind can help you visualize the various stages your tasks undergo from start to finish.
The primary challenge often lies in identifying bottlenecks and issues within the workflow. To address this, it is essential to find a method to measure the time it takes for work to progress from start to finish.
To achieve more precise metrics for decision-making, design a detailed workflow that isolates specific stages of the work’s lifecycle. For example, by distinguishing between active and waiting stages, you can determine whether issues stem from prolonged waiting times or inefficiencies during active phases. Once bottlenecks are identified, you can collaborate with your team to investigate the causes and develop an action plan to improve the situation.
2. Lack of WIP (Work in Progress) Management
Starting tasks without focusing on finishing them and allowing work to age without valid reasons will eventually cause problems. If you let items age, you'll eventually need to close these tickets and might be facing some that are several hundred days old, which can drastically impact your Cycle Times. If 85% of your tasks are completed within 13 days or less, imagine the repercussions of finalizing items that are over 100 days old!
To prevent accumulating too many "old" tickets, discuss them with your team and decide on a course of action, such as moving them back to the backlog for resizing or splitting them up into smaller items. By using Work Item Age as an indicator of potential issues, you can initiate conversations early and address problems before they escalate. While it may not be possible to eliminate all old items, you will likely have significantly fewer on your board. Consequently, if items move faster through your workflow, your Cycle Times will decrease. Additionally, the fewer outliers you have, the more predictable you will become.
3. Having a Blocked Column
This is the worst approach of them all. It doesn’t do what you think it does. You might argue, “But Margaux, it brings visibility to all currently blocked items and helps us manage the blockage better.” (Insert skeptical Margaux face). Yes, it does provide visibility, but does it actually help manage the blockage?
To handle blockages or dependencies effectively, you need to identify where the blockage occurred in your workflow: Was it during the testing phase? The development phase? Or while waiting for UAT? You also need to know how long the item has been blocked because different durations may require different approaches (e.g., escalation).
Moreover, if you have WIP (Work In Progress) limits, what happens when the item is unblocked but no slots are available in the workflow to address it? It will have to gather dust until a slot opens up, extending the time required to handle it.
There are better ways to highlight blocked items without moving them to a different column or status (don’t do that either). For example, in Jira, you can flag an item, which will then change color to make it easily identifiable. This method provides more accurate analytics on time spent in each workflow stage, helping you identify bottlenecks where work often gets stuck. This insight can help you improve your processes and hopefully reduce blocked time.
4. Focusing too much on WIP Limits
WIP limits are an effective way to maintain a balanced workflow, ensuring you neither overload nor underutilize your system. However, they can sometimes mask underlying issues in your process. Consider a scenario where your In Progress stage has a WIP limit of 10, and you adhere strictly to this limit. While you may not exceed or fall short of this number, the age of the items in progress becomes a critical factor. You might be advancing 6-7 items swiftly, but what about the remaining 3 that have stagnated for days? Merely observing WIP limits won't prevent work from aging. To manage your flow effectively, you must monitor both the quantity and age of your WIP. This relates to our third point: actively monitoring and managing the age of your work can resolve numerous issues.
5. Not Slicing Your Work
When work takes too long, it might be because the task is too complex and could benefit from being divided into smaller yet still valuable pieces. What are the smallest deliverable components that still provide value? If a component doesn’t add value, should it even be worked on or released? Having this discussion ensures that you always focus on potential value and avoid discovering at the end of the sprint or project that half of the work was irrelevant. More refined tasks are likely to move smoothly through your workflow, ensuring a steady stream of completed work ready for release. This approach enables you to receive feedback much faster, allowing you to adapt and make better data-driven decisions.
In Conclusion
Let's recap the 5 points and use more familiar terms:
#1: Bad workflow design- ineffective Cycle Time management.
#2: Ignoring WIP Management - overlooking Work Item Age.
#3: Maintaining a blocked column - poor WIP oversight.
#4: Mismanaging WIP limits - inadequate handling of Work Item Age and poor WIP control.
#5: Failing to break down tasks - poor Throughput management.
You might be thinking, "Wait a minute... aren't these the key flow metrics?" And you'd be right! Flow metrics are crucial to achieving better workflow and agility. They're a fundamental concept in Kanban, but they are equally beneficial whether you practice Scrum or SAFe.
Definitions:
Cycle Time: total elapsed time an item took to be completed from start to finish
Work Item Age: total elapsed time since the work started
WIP: amount of work that has started but not yet finished
Throughput: amount of work finished in a given time period
Ultimately, these metrics help you better predict the amount of work that can be planned for the next PI or sprint.
Imagine spending minimal time on metrics while confidently forecasting and achieving accurate predictions. It’s entirely possible! You don't have to choose between methodologies; you can integrate them to achieve your desired outcomes. Shake things up and give it a try 😄
Interested to hear more? Curious but not yet convinced? Feel free to reach out to me at support@55degree.se or on our community at https://community.55degrees.se/.
コメント