A New Copy!

 

Chapter 9 of Daniel S. Vacanti’s Actionable  Actionable Agile Metrics for Predictability: An Introduction (buy a copy today) is titled, “Flow Debt.” The chapter on flow debt caps this section of the book on integrating flow metrics and the cumulative flow diagram.  In this chapter, we use the approximate average cycle time (defined in Chapter 5) to identify flow debt and to guide teams and other stakeholders to ask good questions. Identifying when and where flow debt is being incurred allows teams to take action so that the flow of value can be improved!

Since this chapter focuses on debt, Vacanti begins the chapter with a discussion of three types of debtors from the work of the economist, Hyman Minsky as a metaphor for the types of behavior that encounter and incur flow debt. The three types of debtors are hedge, speculative and Ponzi. Hedge debtors service both the interest and principle of their debt.  If you are paying off a mortgage or a student loan, Minski would classify you as a hedge debtor. Speculative debtors only service the interest on their debt.  Balloon loans generally only service the interest until the end.  Ponzi debtors keep taking out more debt to pay off earlier debt.  Each type of debtor incurs and reacts to flow debt differently.  We recognize flow debt by comparing approximate average cycle time to the actual approximate average cycle time.  This comparison is accomplished by comparing the approximate average cycle time as predicted by the CFD with the actual approximate average cycle time calculated based on the items completing in any given period of time.

Two reminders from previous chapters:

  1. We can use the CFD to visualize approximate average cycle time using CFD Property 4.  CFP Property 4 states that the horizontal distance between the top and bottom line at any point in time is the approximate average cycle time. This property holds between any two flow steps.
  2. Little’s Law Assumption 4 indicates that the average age of work in process (WIP) should neither be increasing nor decreasing.

Based on  the comparison of approximate average cycle time from the CFD to what is actually happening there are three possible scenarios:

  1. The approximate average is greater than actual average 

When the approximate average is greater than the actual average the process is incurring flow debt.  Flow debt is incurred when a team borrows cycle time from some items to get other items done sooner.  The assumptions that allow us to use Little’s Law stipulate that all items once started will eventually complete.  Putting an item on hold to expedite another slows the first one down so it will not be completed in the timeframe suggested by approximate average cycle time which generates flow debt. Flow debt occurs when an item is slowed down (or stopped) so that something else can happen. Similarly, working on items that are eventually thrown out (not completed) cause the same type of flow debt.  These are common reasons for flow debt to occur however, there are all sorts of other ways reasons this scenario can occur.  In order to interpret why flow debt is occurring and whether the situation requires action, whoever is reviewing the metrics needs to ask questions to generate information.

Vacanti uses the three debtor types to show how and why different types of people or teams will generate flow debt. 

  • A hedge debtor will minimize accepting expedited (or throwaway items) work items.  Expedited items are often unplanned items that a team is asked to accept and put ahead of planned work. Their bias is to service the principle and debt; therefore, minimizing the build-up of flow debt is an important behavior.
  • Speculative debtors will always have one (or more) expedited item as work in process or perhaps are violating the WIP limit for expedited items.   The behavior pushes the boundary of what is possible which often disrupts predictability.
  • Ponzi debtors build up flow debt adding more and more items to the expedited queue until no more can be accepted or work jettisoned and the house of cards collapses.  

Interpreting the difference between the approximate average cycle time from the CFD and the cycle time generated from the items actually completing will indicate whether flow debt is being incurred.  What it does not tell the person looking at the data is why.  Using the debtor metaphor provides a structure to narrow the types of questions you would need to ask to help a team discover what is really going on and potential solutions. 

  1. The approximate average is less than the actual average

This scenario generally reflects paying off flow debt and finally completing the items that have been put on hold.  Paying off the flow debt gets the process back towards a stable system.  Until the system reaches stability it will not be predictable. 

  1. The approximate average roughly equal to actual average

This scenario represents a stable and predictable system.

There are all sorts of pressures that can destabilize a system.  All of us have fallen victim to the emergency work item or feature that a stakeholder just has to have  . . . tomorrow.  Each of these scenarios can destabilize the system.  Vacanti caps the chapter by asking the $10,000 question: When is the difference between the CFD and the actual enough to be acted upon?  The answer boils down to the classic answer, “it depends.” In order to know whether we need to take actions starts with understanding context, how big the difference, and the trend.   

 

Previous Installments

Introduction and Game Plan
Week 2: Flow, Flow Metrics, and Predictability
Week 3: The Basics of Flow Metrics
Week 4: An Introduction to Little’s Law
Week 5: Introduction to CFDs
Week 6: Workflow Metrics and CFDs
Week 7: Flow Metrics and CFSs
Week 8: Conservation of Flow, Part I
Week 9: Conservation of Flow, Part II

Support the author (and the blog), buy a copy of Actionable Agile Metrics for Predictability: An Introduction by Daniel S. Vacanti

Book
Kindle

Get your copy and begin reading (or re-reading)!