If you’re a fan of Seinfeld, you may recall an episode where Elaine experiences an alternate reality of her circle of friends. A “Bizarro World” where polar opposites exist. Elaine meets Kevin, Gene, and Feldman who are kind, generous and lacking all the immature qualities of Jerry, George, and Kramer.
For those of us who aren’t familiar with the episode “Bizarro Jerry”, perhaps a takeaway from the story is that there probably is a better version of ourselves out there but, despite our best efforts, we tend to have difficulty recognizing the art of the possible (even when it appears right before our very eyes).
We might use this analogy to bring our attention to a pattern that often appears when analyzing the flow of defects in a value stream. We all realize that high-quality deliverables are crucial to customer satisfaction and we go to great lengths to assure the defects we find during development are resolved quickly. However, we tend to be considerably slower and have much more work in progress (WIP) when resolving defects that are reported by customers via our customer service centers.
The Bizarro Defects Pattern
When you first analyze the end-to-end flow of defects across the value stream, you may be surprised by how much effort is required to manage them efficiently. This is a common phenomenon when the entire value stream set of activities is made visible through Flow Metrics. We often don’t realize how much time and resource is consumed on rework and corrections.
- Flow Distribution indicates a high percentage of defects.
- Flow Load indicates a large amount of defect work in process.
- Flow Time indicates a reasonable overall lead time for defect resolution.
- Flow Velocity indicates a high amount of defect throughput month over month.
- Flow Efficiency indicates unusually high efficiency.
With further analysis, we see this Bizarro Defects pattern emerge. In Figure 2 below, the first set of value stream metrics is the detailed view of the flow of defect resolution for defects reported by users via the service center of the enterprise.
In Figure 3, we see a detailed view of the flow of defect resolution for defects discovered in the development process (either within a sprint or within a QA function) and prior to being released to customers.
By comparing the two dashboards, we can see that it takes considerably longer to fix user-report defects. Most notable is the Flow Load, a metric for WIP. Consider this example: The Flow Load for user-reported defects has a value of 135 defects in progress which is about six weeks of work given the teams’ measured Flow Velocity and Flow Time. Further, it’s clear by the constant volume of WIP that some portion of those defects in progress has undoubtedly stalled or neglected.
On the other hand, we see in this example the Flow Load for in-development defects is nine, which is remarkably good and is directly related to the reason Flow Time is almost 5x faster than for those defects reported by users.
Trusting that this observation is more common than not and that the evidence shows that value stream teams can make defect processing flow exceptionally well for development vs. for customer requests, the question arises, why not focus on making defect correction flow faster for our customers? In other words, why do we stick with Jerry, Kramer, and George when we could be with Kevin, Gene, and Feldman? Well, perhaps Elaine demonstrated it best in showing that we are creatures of habit and it’s difficult to change.
Addressing the Bizarro Defects pattern
Unlike Elaine, our teams need to be customer-centric in order for our enterprises to survive and thrive in this digital era. So, if you discover that your value streams are demonstrating this Bizarro Defects pattern, what might you do to improve the customer experience?
The key is to measure your end-to-end defect flow, focus on the WIP and find where bottlenecks are lurking. Then perform experiments that target the reduction or elimination of these bottlenecks, starting with those that are most impacting flow.
Planview Viz makes this work visible with instant real-time Flow Metrics. It helps us learn to see where we have you have patterns like this. It runs powerful analytics to provide objective data about the state where work is held up and provides continuous feedback on the impact of your improvement experiments.
With this visibility, you can strike a balance between defects pre and post-production to ensure customer satisfaction, and retention is always prioritized.
Register for a Value Stream Management Workshop
We run highly-interactive VSM workshops to help you find patterns like the Bizarro Defect phenomenon. In these sessions, you will:
- Identify and measure value and protection work in software products
- Analyze a baseline of software delivery performance using Flow Metrics
- Pinpoint bottlenecks, reduce waste and improve velocity