Scrum Events: Myths and Facts

I made these videos and articles during my years as an Agile coach, for the teams and students I coached then, and I keep them here as a free archive. If you’re here to learn how Scrum works, the core mechanics still hold. My current work, on how AI is exposing the leadership paradigms this training assumed, is on the blog and in Come Prepared to Die.

Quiz yourself on the Scrum Events with these Myths and Facts. These cover the goal, inputs, outputs, timing, and tips for the various Scrum events and Backlog Refinement. The ideas here extend “textbook” Scrum and so may not exactly match what you may have discovered in your context, but hopefully you find it a useful exercise.

I also recommend my free, printable “Field Guide to Scrum Events.”

This material is licensed CC BY-NC-ND 4.0.

The Sprint

The Sprint can be as short as two days or as long as two months.

Myth.

The Sprint should be a month or less. The vast majority of teams sprint in 1–4 week cycles (weeks are easier to schedule around than months). Two weeks is becoming most common in the industry.

The team must commit in advance to exactly what they will deliver during the Sprint. If the team fails to do so, it’s the team’s fault.

Myth.

Asking the team to commit to delivering a particular set of work limits their ability to react to new learnings during the sprint, threatens their work-life balance, and encourages cutting corners.

The purpose of the Sprint is to complete all of the items on the Sprint Backlog.

Myth.

The purpose of the sprint is to deliver a working, potentially-releasable “Done” increment that accomplishes the Sprint Goal.

The Sprint Goal should be specific, measurable, achievable, relevant, and time-bound.

Fact.

Yes, this is a “SMART” goal. This mnemonic is useful to Product Owners and teams when creating their Sprint Goal. See ap.tips/smart.

The whole Scrum Team must participate throughout the Sprint.

Fact.

Yes, the Product Owner, Scrum Master, and Development team must work together to accomplish the Sprint Goal.

The Scrum Master can cancel the Sprint if the Development Team says the work isn’t possible to accomplish within the Sprint timebox.

Myth.

Only the Product Owner can formally decide to cancel the Sprint. They are most likely to do so when the Sprint Goal is no longer relevant.

The Scrum Team should limit contact with people outside of the team so that they can stay focused on accomplishing the Sprint Goal.

Myth.

Per Agile Principle #4 (ap.tips/principles), “Business people and developers must work together daily throughout the project.” It is all too easy for the Scrum Team to develop the wrong product based on a simple misunderstanding, even with comprehensive documentation.

The Development Team and Product Owner can change the Sprint Backlog mid-sprint.

Fact.

The Development Team and Product Owner can collaborate to clarify and change the Sprint’s scope as more is learned, but the Sprint Goal should be preserved whenever possible.

One-week Sprints have significantly more overhead than two-week or longer Sprints.

Myth.

Although a team conducting one-week sprints would have two Sprint Planning (ap.tips/planning), Sprint Review (ap.tips/review), and Sprint Retrospectives (ap.tips/retro), the Planning and Review events take about 40% less time for one-week sprints because there are fewer Backlog Items to plan and review with stakeholders and end users.

“Sprint 0” or “Design Sprints” in which teams plan out the work to be done for a longer period of time in great detail are a good way of reducing risk.

Myth.

This is simply a traditional, phased-gated fixed planning approach delivered two weeks at a time. Even if it’s just a clickable prototype, try to start building something that you can get feedback on from stakeholders in your first sprint.

Sprint Planning

The three goals of Sprint Planning are: to Identify a Sprint Goal, forecast what Increment can be delivered in the Sprint, and jointly plan how that increment will be achieved.

Fact.

Exactly!

Sprint Planning should take less than an hour per week of Sprint length.

Myth.

It typically takes 1–2 hours per week of Sprint duration to plan a sprint well. If the team cannot complete Sprint Planning within 2 hours per week of Sprint Duration, they should consider performing more Backlog Refinement in advance.

Sprint Planning is optional for the Product Owner as long as there is enough Ready work to begin the sprint.

Myth.

The entire Scrum Team should attend. We need the Product Owner to help craft a good Sprint Goal, clarify the work to be done, and answer questions about whether or not the planned approach is likely to satisfy the customer.

Stakeholders, customers, and other subject matter experts can attend Sprint Planning if invited by the team.

Fact.

It’s not common as much of their feedback will come to the team at Sprint Review, but the Scrum Team is welcome to invite anyone who could help.

Teams should avoid starting or ending Sprints on Mondays and Fridays.

Fact.

Holidays and voluntary time off most often occur on Mondays and Fridays, so you’ll wind up having to reschedule events more often or miss important participants.

The Product Owner dictates the quantity of stories that will be picked up by the Development Team in the sprint.

Myth.

The Product Owner should come to Sprint Planning with a prioritized Product Backlog, but only the Development Team can choose how many of those stories make it into the Sprint Backlog based on their historical performance whenever possible.

The best way to predict what can be completed in the sprint is to break down the work into tasks and assign tasks to developers up to their hourly capacity for the sprint.

Myth.

Absolute time-based estimates are usually less accurate and take longer to identify than relative estimates such as Story Points.

The Development Team should take about half of Sprint Planning to break down backlog items into smaller tasks, each taking a day or less to finish.

Fact.

This is a very important but often overlooked part of Sprint Planning. Tasks are to “Engineering Requirement Documents (ERDs)” that Backlog Items are to “Project Requirement Documents (PRDs).” This discussion creates alignment about the anticipated approach and helps reduce risk.

The team should take a 5–15 minute break every 60–90 minutes.

Fact.

Our attention span lasts about 60–90 minutes at best. Nobody can sit through a 2+ hour meeting and remain productive without a mental and “bio” break.

The team should significantly exceed the forecasted velocity for the upcoming sprint if they have a good “feel” about it.

Myth.

It’s far better to underpromise and overdeliver. If the team’s historical velocity and capacity suggests a forecasted velocity of 30 points this sprint and the team wants to take on 40, try to complete the 30 points early and pull in additional work mid-sprint.

Daily Scrum

The Daily Scrum should happen at the same time and place every day.

Fact.

This consistency makes things much easier. Some teams will cancel the Daily Scrum on the last day of the sprint, but we recommend against that as the team could use this to coordinate their final effort against the Sprint Goal.

The Daily Scrum is primarily a status meeting for stakeholders so they can see the team making progress.

Myth.

While stakeholders and other subject matter experts may always attend, the goal of the Daily Scrum is to inspect and adapt progress against the Sprint Goal, not update stakeholders (including the Product Owner or Scrum Master).

The entire Scrum Team must attend the Daily Scrum.

Myth.

The Scrum Guide suggests that the Daily Scrum is required for the Development Team. The Scrum Master need only ensure that it’s happening and well. On the other hand, most Agile coaches would suggest that the entire Scrum Team should attend and some even suggest that the Product Owner and Scrum Master give updates, too.

It doesn’t matter how the Development Team conducts their Daily Scrum so long as it remains within the 15-minute timebox and meets the same goals.

Fact.

Three questions from earlier versions of the Scrum Guide are commonly used (What did I do yesterday that helped the Development Team meet the Sprint Goal? What will I do today to help the Development Team meet the Sprint Goal? Do I see any impediment that prevents me or the Development Team from meeting the Sprint Goal?). However, the team can use any approach that meets their needs best.

The team, especially distributed teams, can schedule more than 15 minutes for this event.

Fact.

Distributed teams should consider scheduling more than 15 minutes for this event. The Daily Scrum itself is still time-boxed for 15 minutes, but it can be very useful to keep the video conferencing bridge reserved for “Parking Lot” topics. Distributed Scrum Team members are less likely to speak to one another informally throughout the rest of the day.

The Development Team should start the Daily Scrum on time, even if much of the team is missing.

Fact.

Within a few days, attendance should improve as being late to a meeting in progress is more uncomfortable than being late to a meeting that hasn’t started yet.

Anyone, including those outside the team, may raise “Parking Lot” topics.

Fact.

If any Scrum Team member feels that a discussion during the Daily Scrum is taking too long and threatens the time-box, they can suggest further discussion be held for the “Parking Lot.” Anyone who wishes to do so can stick around after the Daily Scrum and discuss these topics: participation is fully voluntary at this point. 

The Development Team should consider using information radiators such as task boards and burnup/burndown charts to augment their Daily Scrum.

Fact.

Task boards, in which all of the tasks are represented in status columns (To Do, In Process, Done) for Backlog Items in rows can significantly improve shared understanding and expedite the standup. Burnup/burndown charts can show the completion rate of backlog items against the ideal rate for completion.

The output of the Daily Scrum is a plan for the next 24 hours and a newly updated Sprint Backlog as necessary.

Fact.

Exactly. This is not a status meeting. It’s an opportunity to inspect and adapt progress against the Sprint Goal.

It’s not a big deal if a member of the Development Team is working on something that they don’t talk about or reflect on the board during at least one team’s Daily Scrum.

Myth.

While the Development Team has a great deal of autonomy over how the Sprint Backlog will be accomplished, they are accountable to the Scrum Team for staying focused on the activities that help the team meet the Sprint Goal. If they truly believe that they should be working on something, they owe the rest of their team transparency.

Sprint Review

A team should claim partial credit for incomplete Backlog Items and re-estimate the carried-over Backlog Items for the new Sprint.

Myth.

Remember Agile Principle #7: “Working software is the primary measure of progress.” If we take credit for partial work, we create the illusion of progress and often wind up starting a lot of work that we don’t finish.

The goal of Sprint Review is to prove progress against a fixed plan to stakeholders.

Myth.

While we certainly want to foster a culture of transparency and trust between the Scrum Team and stakeholders, we also need to elicit feedback on the Increment and collaboratively improve the Product Backlog together.

The primary output of Sprint Review is an updated Product Backlog.

Fact.

The Scrum Team should update the Product Backlog to reflect stakeholder feedback and new knowledge about the product’s context, such as the competitive landscape and the company’s resources.

The input to Sprint Review is the updated Sprint Backlog, Product Increment, and knowledge as to whether or not the Sprint Goal was met.

Fact.

Exactly. We want to be transparent with our stakeholders and show the true state of our work.

Everyone on the Scrum Team and key stakeholders and end users invited by the Product Owner should come to Sprint Review.

Fact.

There’s little value to holding a Sprint Review if stakeholders aren’t present to provide feedback and the business context.

The Development Team should demonstrate all Backlog Items, even if they are incomplete.

Myth.

Don’t demonstrate incomplete Backlog Items at the Sprint Review. It creates the illusion of progress and confusion about what is and isn’t done. Your stakeholders might think something is ready to ship when it isn’t (e.g., looks feature complete but doesn’t pass tests) and ask that you put it into production.

Sprint Review should take 15-20 minutes for a two-week Sprint.

Myth.

It takes about 15-60 minutes to conduct Sprint Review per week of Sprint duration. 1 hour is about right for a 2-week sprint as you need time to demonstrate the work, solicit meaningful feedback, hear updates about the market, and share information about potential release timing.

At the end of Sprint Review, the Scrum Team should discuss the timing, resources, potential features, and distribution channels for the next potential release of the product with stakeholders.

Fact.

This is one of the few times in a Sprint that the Scrum Team, stakeholders, and even end-users will all get together to inspect and adapt. Now that you’re together, take the time to collaborate on the release with one another.

The Development Team should briefly discuss the highs, lows, and blockers encountered during the Sprint and how they were responded to.

Fact.

It can be very helpful for stakeholders to hear about the challenges the team has faced. If a stakeholder hears about a tenacious systemic impediment from multiple teams, they are more likely to believe it exists and act upon it.

Stakeholders should use the Sprint Review to criticize the team if they do not complete all of the items on the Sprint Backlog.

Myth.

This is likely to demoralize the team and incentivize raw output of features over quality. If stakeholders have concerns about the scope of work delivered, they should bring it up with the Scrum Master and/or Product Owner individually.

Sprint Retrospective

An efficient team can conduct an effective Sprint Retrospective in 30-45 minutes.

Myth.

It really does take about 1½ hours to have a meaningful retrospective for a two-week sprint. Cutting Sprint Retrospective too short significantly impacts the team’s ability to inspect, learn, and adapt and impedes continuous improvement.

Anyone can attend the Sprint Retrospective to provide feedback to the team about their processes.

Myth.

Only members of the Scrum Team may attend unless there is consensus from the team that others should be invited.

We should avoid casting blame on individuals.

Fact.

As Norm Kerth says in his Retrospective Prime Directive, “Regardless of what we discover, we understand and truly believe that everyone did the best job they could, given what they knew at the time, their skills and abilities, the resources available, and the situation at hand.” If someone made a mistake, it’s probably because the context they were operating in made it likely. This helps us address the most likely root cause of our challenges.

Two techniques to get at the root cause of an impediment are to ask “why” five times or consider “tools, process, structure, culture.”

Fact.

Asking why five times helps us get at the root cause. Considering “tools, process, structure, culture” also helps: for instance, are we managing dependencies (process) when we could be eliminating them (structure)?

Retrospectives are primarily a feel-good event for teams and can be held less regularly if we are busy.

Myth.

Retrospectives are the most important event because they allow us to consider and improve how we are working together. If we do not retrospect, we are not truly in control of our tools, process, structure, or culture.

The outputs of Sprint Retrospective are new Backlog Items for the next sprint and changes to the Definition of Done, Definition of Ready, and Working Agreement.

Fact.

Some Scrum Masters and coaches call new Backlog Items “actions” and changes to the Definition of Done, Definition of Ready, and Working Agreement “decisions.”

The team should come to Sprint Retrospective prepared with objective data, information, and observations.

Fact.

The Scrum Team should consider The Scrum Team’s progress against the Sprint Goal and Sprint Backlog, the well-being of the people and relationships in and around the Scrum Team, and the effectiveness of the Scrum Team’s tools, process, structure, and culture.

The Sprint Retrospective should occur at the end of the Sprint just before Sprint Planning.

Fact.

It’s ideal to inspect and adapt before Sprint Planning so that we can incorporate Backlog Items related to continuous improvement in our next Sprint. However, every Scrum event is an opportunity to inspect and adapt!

The team should bring snacks and regularly change the Sprint Retrospective format to keep it fun and interesting.

Fact.

Having celebratory snacks can really break the ice and create a friendly and supportive atmosphere, even if you are working remotely. The Scrum Master should occasionally change the format of the Retrospective to bring in new kinds of perspectives and help prevent it from becoming a rote exercise.

Backlog Refinement

Backlog Refinement is an official Scrum event.

Myth.

Although many Scrum Masters and Agile Coaches teach that Backlog Refinement is an event⁠—and it can be quite effective to treat it as one⁠—it’s not officially an event in the Scrum Guide. The Product Owner and Development Team can refine the backlog any way they like and still be true to “textbook” Scrum.

Backlog Refinement can take up to 10% of a Development Team’s capacity, especially at the start of work in a new domain or area.

Fact.

Getting together to talk about the work we expect to complete together is a very good use of time. Every minute spent in the Backlog Refinement activity is at least a minute saved at Sprint Planning, especially because it allows the Scrum Team to get crucial answers from stakeholders and subject matter experts who may not be present before Sprint Planning.

The goal of Backlog Refinement is to detail, improve, estimate, and prioritize Product Backlog items collaboratively to increase shared understanding, reduce risk, and shorten the duration of Sprint Planning.

Fact.

Yep! Although Sprint Planning is the last responsible moment to detail, improve, estimate, and prioritize, doing so early makes the team much more effective and reduces risk.

Every member of the Development Team must participate in Backlog Refinement.

Myth.

Although getting together as a Scrum Team can significantly improve shared understanding of the work, there is no strict requirement for everyone to participate.

The Product Owner should make a sincere attempt at refining and prioritizing the work to be done before coming to Backlog Refinement.

Fact.

Whomever created the Backlog Item should make a sincere effort to elaborate on it before bringing it to the rest of the team for discussion. They should not expect it to survive discussion unchanged, but a little preparation goes a long way.

The team should have a Definition of “Ready,” a list of criteria that indicate when a Backlog Item is likely ready to pull into a sprint.

Fact.

Although the Definition of Ready isn’t an official Scrum concept, it is useful to at least agree that the Backlog Item can be completed within one Sprint.

The Product Owner should send out a list of the Backlog Items to be discussed before the Backlog Refinement event.

Fact.

This allows the Scrum Team to read ahead and familiarize themselves with their description, order, and value. It also helps communicate the full scope of the Backlog Refinement event to encourage brevity. If the Scrum Team gets these stories to “Ready” early it can end the event early.

The Scrum Team can invite other participants, such as stakeholders, subject matter experts, and customers, to Backlog Refinement activities.

Fact.

The Scrum Team can invite anyone they like. It can also be useful for a Product Owner to review proposed backlog items with customers before bringing them to the Development Team.

The ideal output of Backlog Refinement is about 1-½–2 sprints worth of prioritized, understood, and ready backlog items.

Fact.

Unless the team has a very stable velocity and estimates the work during Backlog Refinement, there’s always the risk of the team being able to pull more work into the Sprint at Sprint Planning than the Product Owner has prepared. On the other hand, you don’t want to spend time detailing work too far into the future as you may discover the work isn’t important or your user’s needs were different than you thought.

It’s sufficient to simply average the estimates returned by the Development Team. For instance, if two developers think the Backlog Item is a “1” and two think it’s a “5,” we should just call it a “3.”

Myth.

Creating a smooth flow of work is only half of the benefit we can get from estimating user stories. Suppose the developer suggesting a “1” has discovered a faster way to accomplish the work, or the developer suggesting a “5” is aware of a sticky edge case. If you differ much in your estimates, explain why and revote for another round or two until you converge.