Retrospectives are meetings
A retrospective length is dependent on sprint/iteration length
Retrospective variety ensures team members do not get bored
There are three basic questions in a retrospective
The Scrum Master or Agile Coach must facilitate the retrospective
A retrospective is designed independently of other Scrum/Agile events
1. Retrospectives Are Not Meetings
Retrospectives are not meetings. Retrospectives are events. Why is this important? We know that words have meaning and if you look at the definitions of “meeting” and “event” you will see that a meeting is an assembly of people for discussion or entertainment. An event, on the other hand, is something that occurs in a certain place during a particular interval. Meetings start late and end late, and to be honest, meetings are typically a waste of time. Events are structured, start and end on time, have a purpose, and dare I say it, follow a process or method.
2. Retrospective Length is Independent of Sprint/Iteration Length
The time needed for a retrospective is independent of sprint or iteration length. For example, a four-week sprint does not require a four hour retrospective. A one hour retrospective is enough time for a high-performing team to identify the handful of action items required to improve their future execution, regardless of the sprint length. An average team, on the other hand, may need up to two hours to help them build individual and team retrospective muscle memory.
To maximize the amount of work not done, a team only needs to gather data (what went well/ what didn’t go well) for five events (planning, standups/communication, sprint execution, review, and retrospective) and this can be done in as little as 10 minutes—shorter with a well-practiced team. Analyzing sprint execution (generating insights), conducting a root cause analysis, and developing action items (lessons learned) are where the team needs to spend the majority of the hour.
Consider this: A four-week sprint will have more standups and more execution days than a one-week sprint, but this does not equate to a 4x increase in the amount of time needed to gather data in a retrospective. A high-performing team should be able to glean all their learning points from a 2-6 week sprint within the span of an hour. Anything more is a disrespectful waste of the organization’s money and the team members’ time.
3. Retrospective Variety is NOT the Spice of Organizational Life
Boredom: “an aversive state of wanting, but being unable, to engage in satisfying activity.”
The idea of trying out a new retrospective activity (game) so team members do not get bored is misguided. The perceived or actual boredom displayed by individuals or teams toward retrospectives may be (1) a result of the type of work the team is doing, (2) a lack of understanding of the retrospective activity and/or a failure to achieve actionable outcomes, or (3) the fact that great retrospectives are hard as they challenge people to be open and honest. We also know that boredom is a result of being too challenged.
Let’s assume that knowledge workers are challenged each sprint or iteration and their work is not the source of the perceived boredom. How does changing the retrospective activity (game) address the averse state of wanting to be engaged in a satisfying activity? And if retrospectives are perceived as being too challenging, then why would one believe that changing the activity would pacify this root cause of boredom?
Consider these points if you continue to believe changing up retrospectives in the name of boredom is a smart idea:
Inconsistent retrospectives, in structure and/or frequency, lead to mediocrity. According to Jim Collins, “The signature of mediocrity is not an unwillingness to change. The signature of mediocrity is chronic inconsistency .”
Repetition. Repetition. Repetition. Learning something new, awkward, and difficult requires some repetition. It takes a while to get used to opening up, being self-critical, and learning how to respectfully challenge each other in and out of a retrospective.
According to Doug Sundheim, “Debriefing (Retrospective) is a structured learning process designed to continuously evolve plans while they’re being executed.”
Culture is a product of retrospection. Edgar H. Schein points out that “culture is the result of a complex group learning process .”
Innovation is a product of interactions, and common processes that accelerate those interactions are necessary. According to a recent McKinsey Quarterly article, “…innovation is a complex, company-wide endeavor, it requires a set of crosscutting practices and processes to structure, organize and control it .”
So not only will “variety” in your retrospectives not produce desired results, it will also have a negative impact on your team’s ability to improve.
4. The Two Most Powerful Questions in a Retrospective Are:
What was the primary objective?
Did we achieve what we set out to do? Did we achieve that objective?
Why are these the most powerful questions during a retrospective? Simple: alignment. If the team is not aligned, if individuals cannot succinctly restate the sprint objective, then we (organization, leaders, Agile Coach, Product owner, etc.) have failed to establish the necessary conditions for the team to become empowered .
With an aligned team, the answer to the second question should be binary (yes/no) assuming the primary Sprint objective is clear, measurable, and achievable. Moreover, the primary Sprint objective should be connected to the external effects on the economic system (i.e. business value) and not an internal measures of performance (e.g. ready stories, velocity, burndown, etc.).
What went well? What didn’t go well? What can we do better? These three questions are commonly used and thought of as a proven retrospective design or process instead of as techniques that are helpful in gathering data and deciding what to do. Caution, a real danger exists when asking a team “What they can do better?” if they do not share a common picture of “What” happened, understand “How” that something happened and, more importantly, “Why” it happened (root cause analysis). In a later post I will go into more detail on how a self-similar activity used in strategy and product development (What-How-Why) can be a powerful tool in building action items in retrospectives.
5. Who Facilitates?
A lot of people look to the ScrumMaster or Agile Coach to facilitate a retrospective, and indeed, when a team is just learning how to conduct and participate in a retrospective, this is important. However, the retrospective should be viewed as a leadership episode, where leadership is valued over facilitating the event. In complex systems, leadership “takes place during interactions among [team members] when those interactions lead to change in the way those [members] expect to relate to one another in the future .” How members relate to each other in the future is directly connected to how a leader (a person in an actual or perceived position of some authority) establishes a safe environment during the retrospective (a leadership episode).
The Product Owner’s position on the team is better suited for establishing this safe environment as they generally have over 51% of the vote when it comes to the product vision and backlog. On the other hand, the ScrumMaster or Agile Coach are often contractors or viewed as event facilitators, protectors of the framework, thus making the crucial step of setting the tone (establishing psychological safety) problematic. Development team members are also great candidates for leading the retrospective and should be given the leadership opportunity when the team is comfortable with a proven retrospective process. It is perfectly acceptable for the ScrumMaster or Agile Coach to coach whomever is leading the retrospective during the actual retrospective execution and then follow up with a one-on-one retrospective using a self-similar process.
6. The Leadership Pattern within a Retrospective Should be Tightly Coupled with Leadership Patterns of Other Events
Iteration or Scrum events should be viewed as a whole and not as standalone parts of a system. Just as Scrum exists in its entirety, the leadership patterns of each Scrum event should be tightly coupled (have interdependencies) where the relationships among those patterns work together in some fashion. Combining an off-the-shelf retrospective game with an ad hoc planning process (the letting the team figure it out approach), for example, is counter to basic Systems Thinking. Leadership patterns within system events need to be connected. How a team plans, how it conducts its standups, and how it executes its sprint must be connected to how the team holds its retrospective.
To illustrate this point, consider Russell Ackoff’s example of taking the best parts from 450+ different cars (best transmission, best tires, brakes, cooling systems, belts, spark plugs, etc.,) and trying to put those best parts together to make the best car. The parts do not connect and therefore you do not have a car, you have a mess. The proliferation of retrospective games, ones that are designed independently of other Scrum events, are the equivalent of those incompatible best parts—they only contribute to an Agile mess.
Brian “Ponch” Rivera is a recovering Naval Aviator and Commander in the U.S. Navy Reserve. He is the co-founder of AGLX Consulting, LLC.
 Jim Collins, Good to Great: Why Some Companies Make the Leap…and Others Don’t (HarperCollins, 2001)
 Edgar H. Schein, Organizational Culture and Leadership, 3rd Ed. (Jossey-Bass, 2004)
 Marc de Jong, Nathan Marston, and Erik Roth, The Eight Essentials of Innovation. McKinsey Quarterly April 2015
 Senge, P. M. (1990). The fifth discipline: The art and practice of the learning organization. New York: Doubleday/Currency.
 Hazy, J. Goldstein, J, & Lichtenstein, B.B. (2007). Complex Systems Leadership Theory: New Perspectives form Complexity Science on Social and Organizational Effectiveness. http://emergentpublications.com/documents/9780979168864_contents.pdf?AspxAutoDetectCookieSupport=1
Agile (c) Can Stock Photo