Cover of Actionable Agile Metircs

A New Copy!

The Oscars are being announced March 4, 2018. At some point in the process, someone will call for the envelope and boom: the big reveal. So in the spirit of the moment, we have our next book for the Re-read Saturday. We will begin re-reading Turn the Ship Around by L. David Marquet in two weeks. I have interviewed Mr. Marquet twice for the podcast; I have gotten a ton from his books and the interviews, so I am pumped. I also want to announce the book after that, The Checklist Manifesto by Atul Gawande (I recently bought a copy and want to share what I have gotten out of it). Now on with the main attraction!

The title of Chapter 16 is Getting Started, in Daniel S. Vacanti’s Actionable Agile Metrics for Predictability: An Introduction (buy a copy today). The never-ending journey of process improvement needs to begin at the beginning. In this chapter, Vacanti lays out an outline for adopting a process improvement approach that uses the metrics discussed earlier in the book.

Step 1 Define Your Process

  1. Begin by defining your process. This is a far less trivial step than it sounds. Defining your process often begins with defining the boundaries which allow a crisp understanding of when work arrives and leaves. This understanding is important for all cycle time metrics. Having led and participated in many process definition events, I know that until you define the boundaries you will end up with all sorts of qualifiers for how and when work enters the process. Listen for squishy words such as except, but, sometimes and the classic, it depends.
  2. Decide on the what counts as work in process. Vacanti argues that the definition is up to the reader. Apply the data once defined. Organizations I have recently worked with have settled on stories, defects, mortgage applications and network connections (four different organizations). My counsel is to consider what needs to be tracked and what is important to the bigger organization.
  3. Review your policies to determine how many violate Little’s Law. Every violation of Little’s Law drags the team further away from predictability.

Step 2: Data Time (otherwise known as Capturing Data)

Chapter 4 walked through an approach to recording work as it enters the boundary of the process and then as it transitions between active steps. Understanding when work transitions between steps or states, generate more information and is the basis for constructing the cumulative flow diagrams outlined in earlier chapters. The richer the data you collect the more information can mined. Richer data means collecting extra attributes about the work being delivered. Extra attributes include who request the work, the application impact and the context that drove the work.

Step 3: How much data do you need to act?

Vacanti calls out Douglas Hubbard‘s rule of five (How to Measure Anything – an earlier re-read). The rule of five indicates that there is a 93.75% chance that the mean of a population is between the smallest and largest values in any random sample of five items. Bottomline, the need for huge quantities of data is fairly low. A CIO early in my career pointed out that using data even if it was not perfect put a lot of pressure on everyone to make it better. That said, the value of your data is only as good as the process to capture and record the data.

Vacanti concludes the chapter with a few reminders about cumulative flow diagrams. He uses Cumulative Flow Diagram Properties One and Two to drive a conversation about how cycle time metrics and cumulative flow diagrams. Property One reminds us that the top line in the diagram represents the items entering the process and the bottom line are those that are leaving. Property Two indicates that no line inside the CFD can go down. Breaking these rules suggests problems both understanding what the CFD is and how do use the CFD.

In summary, you could boil this chapter down to the statement, “define your process, get some data, and then do something with it.” Unfortunately, like many bumper sticker statements, the devil is in the details. In reality, at least 50% of getting started is breaking the inertia generated by how you work today. The rest is recognizing that you need to embrace a process of learning and adjusting based on what you learned. Don’t expect everything to be perfect right out of the gate.

Next week we will summarize the reread of the book. I’m going to leave the key study to you to read by yourself, and then in two weeks, we will start a new book. Next week also, on the podcast, I will publish my interview with the author.

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

Week 10: Flow Debt

Week 11: Introduction to Cycle Time Scatterplots

Week 12: Cycle Time Histograms

Week 13: Interpreting Cycle Time Scatterplots

Week 14: Service Level Agreements

Week 15: Pull Policies

Week 16: Introduction to Forecasting

Week 17: Monte Carlo Method Introduction