Wednesday, February 19, 2025

Flow Metric 7: Flow Turbulence


Lean Principle 3: Make value flow without interruptions.

SAFe® Lean Agile Principle 6 – Make Value Flow without Interruptions

In today’s fast-paced delivery environments, businesses that thrive prioritize flow of value (for purposes of this paper, I am using “flow” to designate the flow of a Feature through the a) ART Kanban or b) through the ARTs delivery pipeline).  ensuring work flows smoothly through the value stream without unnecessary interruptions.    While the existing Scaled Agile Framework Flow Metrics such as Flow Velocity, Flow Time, and Flow Load provide valuable insights, this paper introduces a seventh flow metric: Flow Turbulence.

Flow Turbulence measures disruptions caused by low Percent Complete and Accurate (%C&A) rates across the value stream.  %C&A has been used as First Pass Yield in lean for the past 60 years and in the world of engineering and DevOps since at least 2011 (Humble & Farley)

If overlooked, Flow Turbulence can ripple through upstream and downstream processes, creating delays, rework, and dependencies that can impact the entire system’s ability to deliver value. Let’s explore this metric and the new Scaled Agile Flow Accelerator to address it: Design the System for Flow.


Understanding Flow Turbulence: The Turtle Problem

Imagine a turtle swimming upstream in a flowing river. As it pushes forward, its movements create ripples that extend all the way to the shoreline, disturbing the water’s natural flow. These ripples represent the turbulence caused by low %C&A at any step of the value stream. Just as the turtle’s ripples can disrupt an entire ecosystem, defects and incomplete work disrupt the smooth progression of work in an enterprise flow system.

 


The goal of Flow Turbulence is to measure and reduce these disruptions by understanding how %C&A performance at various points causes rework, bottlenecks, and delays.

What is Percent Complete and Accurate (%C&A)?

%C&A assesses whether outputs of a step in the value stream are delivered in a state that is both:

  • Complete: The task or item is fully finished with all necessary information and components.  Everything is there to successfully act on this value when pulled to the next step.
  • Accurate: The output meets the quality standards required for the next stage without requiring corrections.  For each step in a value stream, %C&A tells us how often work flows successfully through the step – e.g. no defects and no rework

When %C&A is low, two major disruptions occur:

  1. Upstream Disruptions: Work with errors or missing details must be reinjected into the flow for rework, creating bottlenecks and delays.
  2. Downstream Disruptions: Tasks passed along without being complete and accurate lead to missed dependencies, delays, and quality issues in later stages.

%C&A has is origins in Lean Thinking and Lean Six Sigma, with key examples from Toyota by emphasizing the reduction of defects and rework through built-in quality (jidoka),


How Flow Turbulence Affects the Value Stream

Upstream Turbulence

When an incomplete feature or deliverable requires rework, it clogs the upstream workflow. Teams must pause their current tasks to fix errors, causing delays and disrupting their planned capacity.

Example: A product design team delivers specifications that are incomplete, forcing downstream developers to pause, wait for clarification, or make assumptions. As the issue flows back upstream for resolution, it creates a backlog of unfinished work.

Downstream Turbulence

When low-quality outputs are passed downstream, they create cascading effects that disrupt schedules, cause dependency failures, and increase delivery risks.

Example: A partially tested software component causes bugs to propagate downstream to testing and deployment stages, where it requires costly remediation and impacts delivery timelines.

Both types of turbulence result in delays, inefficiencies, and reduced predictability, emphasizing the need to optimize %C&A.


Flow Accelerator: Design the System for Flow

To address Flow Turbulence, organizations must proactively design systems and value streams that prioritize quality at every step. This concept builds on Lean principles, similar to the andon cord approach used in manufacturing to immediately flag quality issues.

Key Practices for Designing the System for Quality:

  1. Integrate Quality Checks in the Flow: Embed checkpoints within the workflow to validate the work is complete and accurate before advancing to the next stage. This ensures defects are caught early, avoiding downstream disruptions.
  2. Implement the Equivalency of an Andon Cord: Just like in manufacturing, where a worker can pull the andon cord to stop production when a defect is detected, teams should have the authority and mechanisms to pause the flow and address issues before they escalate.  SAFe Principle 9: Decentralize Decision Making is vital to success in this step.
  3. Introduce a 'Pause State' for Reflection: Build a systematic approach for pausing and reflecting at key intervals within the value stream. During these pauses, teams assess %C&A, identify the root causes of turbulence, and implement improvements.  SAFe cadenced events such as the Inspect and Adapt Workshop and Team Retrospectives are great examples of incorporating a ‘Pause’ state.
  4. Optimize Feedback Loops: Rapid feedback from upstream and downstream teams ensures that issues can be corrected before they cause significant disruptions. Encourage experimentation and continuous improvement through feedback cycles.  Remember that feedback is already present, we just need to improve our ‘receptors’ ability to gather and use feedback.

Improving %C&A with Systematic Quality Practices

  • Cross-Functional Alignment: Ensure that all stakeholders have a shared understanding of what “complete and accurate” means for each stage of the value stream.
  • Automated Testing and Validation: Use continuous integration, continuous testing, and automation of those tests to verify that work meets quality standards at each step.
  • Root Cause Analysis: Regularly conduct retrospectives to identify why %C&A is low and implement improvements.
  • Have a clear agreement on Definition of Ready and Definition of Done: Encourage ongoing discussions on DoR and DoD and continue to optimize these components.

Measuring Flow Turbulence

To quantify Flow Turbulence, organizations can track:

  • The percentage of stories, tasks or features requiring rework due to low %C&A.
  • The frequency and delay time of disruptions reported at downstream stages.
  • The cumulative delay caused by rework and defects.

Visualizations such as cumulative flow diagrams (CFD) or scatter plots of rework incidents can help identify where turbulence originates and its impact on delivery timelines.


Benefits of Managing Flow Turbulence

  • Faster Delivery: By reducing rework and dependencies, teams can deliver value more predictably and quickly.
  • Improved Quality: Higher %C&A ensures downstream teams receive high-quality inputs, leading to fewer defects and better outcomes, preventing the ‘good work on top of bad’ problem.
  • Enhanced Collaboration: Teams communicate and collaborate better when there are clear quality standards and feedback mechanisms.

Conclusion

Flow Turbulence, driven by low %C&A, is a critical indicator of disruptions within a value stream. Addressing this turbulence through the Design the System for Quality Flow Accelerator ensures that organizations can reduce rework, improve flow efficiency, and enhance overall business agility. Just as the ripples from a turtle swimming upstream can disrupt an entire riverbank, unchecked turbulence can have cascading effects on delivery outcomes.

By building quality into every step of the value stream and embracing reflective pauses and rapid feedback, organizations can swim upstream smoothly, delivering high-quality outcomes without the turbulence.

 

# First Pass Yield (measuring quality at each step), and by measuring the quality of handoffs between stages (Six Sigma).  (FPY = (Number of good units + Number of acceptable units) / Total units entering the process × 100%)

Andon Chord - means "STOP". The work needs to be fixed before it moves forward. In a CDP this might translate into "all tests run green" before the work gets pulled into the next step.

No comments:

Post a Comment

Flow Metric 7: Flow Turbulence

Lean Principle 3: Make value flow without interruptions. SAFe® Lean Agile Principle 6 – Make Value Flow without Interruptions In tod...