Editorialized Glossary

A glossary, with my editorial thoughts.

  • Agile: An approach to knowledge work that is best described in the four value statements and 12 principles of the Manifesto for Agile Software Development. I consider agility as a direction rather than a destination, such that questions that assume a binary answer like “Are you agile?” are inappropriate and often impede productive discussion. Given the ubiquity of the word today, if pressed beyond the manifesto, I tend to prefer meaningful and descriptive words (rather than describing something as “agile”) like “iterative and incremental delivery.” (I try to avoid the word waterfall for the same reason.)
  • Agile Industrial Complex: See Agile Industrial Complex.
  • Anzeneering: Inspired by Josh Kerievsky’s portmanteau combining anzen (Japanese word for “safety”) and engineering, Anzeneering is “safety engineering.” Leaders at all levels should be actively working to create psychological safety in teams and throughout the organization, since it is the foundation for innovation and high performance. (See Psychological Safety and Anzeneering)
  • “Big Three” Flow metrics: Throughput, Delivery Time, Work in Progress
  • Blocker clustering: A flow-analysis technique that uses records of issues that have blocked work items, grouping them by common cause. Blocker clustering was originally promoted by Klaus Leopold and Troy Magennis; it is a way to learn from blockers and quantify their impact on your delivery system. Data helps point toward system improvements. (see https://www.slideshare.net/MatthewPhilip/flow-from-blockers-how-to-use-blocker-clustering-to-improve-predictability-ale-2015-conference)
  • Capability: What you can do (organization, team, individual)
  • Capacity: How much you can do (organization, team, individual). Best forecast probabilistically; I strongly advise against the SAFe-prescribed method.
  • Change-fail rate: https://www.thoughtworks.com/radar/techniques/four-key-metrics
  • Chapter: A group of people currently in the same functional role (e.g., QAs, Developers, Scrum Masters). A bounded rather than centered group. Popularized by the so-called Spotify Model (see Spotify Model). Here’s my presentation on Chapter Leadership.
  • Class of Service: A combination of policies and cost-of-delay archetypes that inform how certain types of work items are treated. These categories of work items may warrant different policies for selection and processing based on different customer expectations, relative value, risk tolerance and/or cost of delay. Example: Pull a low-risk tolerance item when it is within its 95th-percentile confidence interval of being delivered.
  • Coaching: One of the professional Scrum “developing people and teams” competencies (https://www.scrum.org/professional-scrum-competencies/developing-people-and-teams)
  • Collective product ownership: Rather than a single Product Owner
  • Commitment point: The point in a kanban system at which someone commits to deliver a work item, which is to say that it is no longer managed as merely an option. Before this point, work done supports the decision whether or not to deliver the item. After this point, it has been confirmed that the customer wants and will take delivery of the item, and that the service will deliver it.
  • Complexity: A way of thinking about the world, and an under-appreciated aspect of knowledge work that bears on everything from strategy to estimating work. Appreciating complexity (see Cynefin) is an imperative for anyone who coaches or otherwise intends to lead in intangible-goods systems. As Stephen Bungay writes in The Art of Action: “Organizations … are complex adaptive systems trying to survive and prosper in a fitness landscape full of diverse organisms with different agendas in which their interaction produces unpredictable first-, second-, and third-order effects. Every cause is itself an effect and every effect a cause, linked by feedback loops, some dampening, some reinforcing. Changing an organization involves careful judgment about how and where to intervene in the system.”
  • Connectivity Index: From Mik Kersten’s Project to Product, this measures accidental complexity (the ratio of repositories in the tool network that have been connected to those that have not).
  • Core Protocols: The Core Protocols have been around a while. They’re newly relevant because of a renewed need in this quarantine season for teams to work well together through humanizing agreements.
  • Cost of Delay: A measurement that considers the value of something relative to time, which makes it a superior measurement to simple value. Technically, it is the difference between the benefit (net present value) that would be available from a work item if it were completed without delay and the benefit if it were delayed by a period of time. Measured in consistent units of value such as dollars.
  • Cost-of-Delay planning: “If you only quantify one thing, quantify the cost of delay” says the paragon of product flow Don Reinertsen. The whole point is to do it in money (economic) terms, which means don’t do it the way SAFe tells you to. (Read more.)
  • Cost-of-Delay Archetype: A way to quickly identify a work item to aid in the Three S’s:
    • Expedite: “Causing us pain now”
    • Fixed-Date: “Has a true deadline.” The key is: We don’t realize any economic benefit from delivering this before the date.
    • Standard Urgency: “The sooner, the better”
    • Intangible: “Time and value are both difficult to assess, but they are nonetheless real and relevant.”
    • Intangible Fixed Date: One that I made up that combines elements of Intangible and Fixed Date.
  • Cycle Time: Elapsed time from the beginning of one work stage to the next (e.g., “The QA cycle time is outstripping the Dev cycle time”). Because of this rather technical definition, I prefer to use Delivery Time (below) to refer to what many people call Cycle Time.
  • Delivery Time: (Aka Flow time, cycle time) Elapsed time from Commitment point to Delivery point. The metric customers/consumers actually care about when they ask when something will be done.
  • Deployment Frequency: One of the so-called “four key metrics” of devops. Measures the number of deployments per period of time (e.g., deployments per week). https://www.thoughtworks.com/radar/techniques/four-key-metrics
  • Dojo Coaching: Whether a physical or virtual space, a dojo provides a safe setting for practice. https://www.solutionsiq.com/wp-content/uploads/2017/08/Agile-Dojo-July2017.pdf
  • Engagement, or Employee Engagement: the levels of enthusiasm and connection employees have with their organization. It’s a measure of how motivated people are to put in extra effort for their organization, and a sign of how committed they are to staying there (https://academy.cultureamp.com/hc/en-us/articles/204539759-What-is-Employee-Engagement-). I prefer this to “employee happiness” because it’s a richer, less variable measurement and to “employee retention” because it is richer and a better leading indicator.
  • (Points-based) Estimating: As evidence mounts about the low correlation between upfront estimates and actual delivery time, this practice often does more harm than good. See Probabilistic Forecasting.
  • Event Storming: I first learned about this “workshop-based method to quickly find out what is happening in the domain of a software program” from Alberto Brandolini, so he can tell you everything you need to know: https://www.eventstorming.com/
  • Experiment design: Hypothesis-Driven Development is thinking about the development of new ideas, products and services – even organizational change – as a series of experiments to determine whether an expected outcome will be achieved, so we need to know how to design and run experiments properly. Improve with experiments As Barry O’Reilly says, “We do not do projects anymore, only experiments.” Experiments have hypotheses, measurements and validated learning that is shared. https://www.slideshare.net/MatthewPhilip/no-lab-jacket-required-designing-experiments-for-learning-xp2020-conference
  • Facilitation: One of the professional Scrum “developing people and teams” competencies (https://www.scrum.org/professional-scrum-competencies/developing-people-and-teams)
  • Feature and story backlogs: See Options.
  • Fitness-for-purpose: Rather than one-size-fits-all frameworks and decontextualized maturity models, fitness for purpose is a powerful concept for organizational improvement and survival, as it focuses on understanding why your customers choose you, your products, and your services. For anyone wondering how to measure “agile adoption,” consider instead the concept of becoming fit for purpose.
  • Flight Levels: Also known as multi-level kanban. Helps create organizational alignment, flow across all levels. (https://mattphilip.wordpress.com/2020/01/31/flight-levels-and-metrics/)
  • Flow:  the movement and delivery of customer value through a process (Vacanti in Actionable Agile Metrics for Predictability)
  • Flow efficiency: Quick, what is the flow efficiency of your value stream? If you’re anything like industry standard — between 5-15% — you probably should focus your improvement efforts on your system’s waitflow.
  • Flow management: Less Scrum Mastery, more flow management, please. From Essential Kanban Condensed: “The flow of work in a kanban system should maximize the delivery of value, minimize lead times, and be as smooth (i.e. predictable) as possible. These are sometimes conflicting goals and, since the deliverables are usually complex, empirical control through transparency, inspection, and adaption is required.”
  • Flow Manager: an emergent role that looks after and encourages flow in a team (see https://www.infoq.com/articles/flow-manager-deliver-fast-smooth/)
  • Flow-Planning meeting: Also known as Kanban Meeting, a different way to practice the traditional standup meeting to focus the conversation on creating flow for the day. Avoids the rote practice of the three questions and instead focuses on the “new three questions”
  • Guild: A community of interest. A centered rather than bounded group. Popularized by the so-called Spotify Model (see Spotify Model). One powerful example that I have initiated and seen work in organization is a guild of facilitators, comprised of people from many different roles who offer the service of facilitation to colleagues.
  • Iterations (or sprints) as checkpoints (not planning boxes): To promote flow-based delivery,  treat iterations (or sprints) as checkpoints not planning boxes, which tend to suboptimize delivery (i.e., optimize for the box, not flow or value). Demo what you have when the cadence occurs, continuously plan.
  • Kanban: A word that has multiple connotations, which is important to understand before entering the conversation:
    • Work-visualization board, derived from the Japanese word “kanban” (meaning “sign,” “signal card,” “tally,” or “large visual board”)
    • A pull-system enabled by virtual kanbans which signal capacity and create flow, reduce overburdening and yield other benefits.
    • Continuous-flow-based delivery; “iteration-decoupled delivery cadences.”
    • Kanban Method: a method for defining, managing, and improving services that deliver knowledge work, such as professional services, creative endeavors, and the design of both physical and software products.
  • Limiting WIP: The key to operating a pull/kanban system. Start with a conWIP (constant WIP across all in-progress stages) and only further divide into stage WIP if necessary. (https://mattphilip.wordpress.com/2020/03/13/do-you-even-know-what-a-kanban-is/)
  • Making policies explicit: “Explicit policies are a way of articulating and defining a process that goes beyond the workflow definition. A process expressed as workflow and policies creates constraints on action, is empowering within the constraints, and results in emergent characteristics that can be tuned by experiment”” (Essential Kanban Condensed). Your kanban board is incomplete without them.”
  • Making work visible: The first practice of Kanban Method. But surprisingly few organizations make their work above the team level visible. Use in conjunction with Flight Levels.
  • Mean time to restore (MTTR) https://www.thoughtworks.com/radar/techniques/four-key-metrics
  • NoEstimates/BeyondEstimates: An approach to answering the question “When will it be done?” using less effort and considering more than simply effort as a source of variation. The approach itself can be described as a spectrum (see https://www.slideshare.net/MatthewPhilip/forecasting-with-less-effort-and-more-accuracy-agile-camp-ny-2018)
  • Objectives and Key Results (OKRs): One of the best ways to create aligned autonomy in an organization (whatmatters.com)
  • On-time Delivery Rate: Measures the rate at which work items with true due dates are delivered by their due dates (e.g., 92% of items).
  • Organizational Change Management: The rate of change requires thoughtful, respectful approaches. Not only is this an individual competency, it is also the hallmark competency of an organization.
  • (Real) Options thinking: Contra the traditional backlog, managing work — whether product-development features or organizational-improvement ideas — as options allows us not only to not feel obligated to do them but also enables queuing discipline related to economics. https://medium.com/@fernando.a.cuenca/commitment-real-options-kanban-392e0ad12a2f
  • Pomodoro Technique: Not a new self-management practice, but one that has gained renewed relevance in our work-from-home period. See https://www.techrepublic.com/article/using-the-pomodoro-technique-to-become-your-own-boss/ and https://www.nytimes.com/2020/06/23/magazine/pomodoro-technique.html for more.
  • Probabilistic Forecasting: An alternative to traditional upfront estimating that provides a range of possible outcomes and a percentage likelihood of each (usually done via Monte Carlo simulation). This approach is enabled by a kanban system. See also NoEstimates.
  • Product backlog management: Given that large backlogs are wasteful and that we want to aim for outcomes over outputs in a VUCA environment, I prefer the practice of Sequencing, Selecting and Scheduling.
  • Retrospective: Better to not to them at all than to go through the motions. But still the most powerful feedback loop for kaizen. Try them at leadership/management levels (see Flight Levels).
  • Roadmapping: To quote Marty Cagan, bad product teams “just build what’s on the roadmap, and are satisfied with meeting dates and ensuring quality” whereas good product teams “know that many of their favorite ideas won’t end up working for customers, and even the ones that could will need several iterations to get to the point where they provide the desired outcome.”
  • SAFe: Before implementing SAFe, read these “Agile at scale generative principles.” My view is that SAFe is a big tool box, so it’s my professional responsibility to help people learn how to use those tools properly.
  • Safety (also Psychological Safety): The extent to which a person or team views the social climate as conducive to interpersonal risk (A. Edmondson). (See Psychological Safety and Anzeneering)
  • Scaling: See Unscaling
  • Scrum of Scrums: Meant to be a synchronization meeting between scrum masters. I encourage scrum masters to sync on an as-needed basis rather than either wait for a scheduled meeting or attend a scheduled meeting when they don’t have a need.
  • Self-Management: Brought into the forefront by our current world situation, one of the foundational competencies of remote work. Includes practices such as deep/focused work and personal kanban.
  • Sequencing, Selecting and Scheduling: Rather than stack-ranked “prioritization,” the “three S’s” work with economic concepts like cost of delay and risk assessment to enable more effective planning.
  • Service-Delivery Review: Kanban Method feedback loop that I’ve called “the missing agile feedback loop”
  • Skill liquidity management: Practice that helps solve problem of how to have the right skills available to match the market’s volatility (https://www.squirrelnorth.com/post/why-are-the-right-skills-never-available-when-you-need-them)
  • Spotify Model: A way of organizing horizontally and vertically such that a person’s primary identity is with his or her delivery team, as opposed to a functional/role affiliation. However, with guilds and chapters, the model provides support that is often missing in strictly delivery-team oriented affiliations. The model comes with the obligatory caveats to not simply install it, be aware of your own organization’s context, etc., though I feel some of the recent backlash is overwrought. It’s a useful model for most groups that I’ve worked with.
  • Standup Meeting: An energized, daily microplanning of a team to plan flow for the next 24 hours, no more and no less. Should have the attributes of a timeout in a sporting contest, like in a basketball game. Should follow a pattern of “walking the wall” in which the team collaboratively views their kanban board and considers the new three questions.
  • Strategy Deployment: Or Hoshin Kanri, if you prefer Japanese. Karl Scotland describes it best: the literal translation… is “Direction Management”, which suggests both setting direction and steering towards it. A more metaphorical translation is “Ship in a storm going in the right direction.”
  • Swim lane: a horizontal lane on a kanban board crossing two or more columns along which cards flow (Kanban Condensed). Not to be confused with Stages, which are the vertical areas on the board.
  • Systems Thinking and Sensemaking: VUCA (Volatile, Uncertain, Complex, Ambiguous) environments require systems thinking. In particular, Cynefin as a thinking tool provides useful decision-making contexts or “domains” that help leaders identify how they perceive situations and make sense of their own and other people’s behavior.
  • Throughput: The number of work items exiting a system or subsystem per unit of time, whether completed or discarded. Measured in: work items per unit of time (Kanban Condensed). E.g., user stories per week. The main service-delivery metric measuring output and one of the “Big Three” flow metrics. A common fitness criterion for customers.
  • Three S’s: Scheduling, selecting and sequencing of work. I prefer this language to simple “priority” because it encompasses cost of delay thinking and planning and provides better nuance.
  • Total Motivation: “Total Motivation is a variant of employee engagement. Total motivation, or ToMo, is the simple theory that why people work determines how well they work. There are six reasons why people work – three lead to higher performance and three lead to lower performance. Researches found that “”a high-performing culture maximizes the play, purpose, and potential felt by its people, and minimizes the emotional pressure, economic pressure, and inertia.”” The six main reasons people work are: play, purpose, potential, emotional pressure, economic pressure, and inertia. https://www.vegafactor.com/survey has some excellent free surveys for leaders, individuals and teams.
  • Unscaling/Descaling: See Andy Carmichael’s seminal post, and more recently Barry O’Reilly’s. And my former teammate at ThoughtWorks, Luca Minudel, presented his Generative scaling principles at the recent XP 2020 conf, a must-read (https://www.smharter.com/blog/agile-at-scale-generative-principles/)
  • User Stories: Are stories inhibiting agility? Read my post: https://mattphilip.wordpress.com/2020/07/21/stop-writing-stories-start-validating-working-software/
  • Weighted-Shortest Job First: With recent insights from the likes of Dan Vacanti calling into question the knowableness of value, not to mention duration, WSJF has limited application to complicated environments (but not complex ones)
  • Work in Progress (WIP): Work that has been started but not yet finished. Work In Progress is one of the most powerful levers for creating speed, predictability and quality, and yet one that is sorely overlooked at all levels of organizations.
  • Velocity: Premised on reliability of Story Points (see Estimating), Velocity is a proxy metric for Throughput. Why not simply use Throughput?