The Sprint

PSM I PSM I · Scrum Events

By Abu Anns · Last updated 20 November 2026Exam version: PSM I (Scrum Guide 2020)

Sprints are the heartbeat of Scrum — the fixed-length events in which ideas are turned into value. A Sprint is one month or less, and its length stays consistent throughout the development effort. A new Sprint starts immediately after the conclusion of the previous Sprint; there is no gap, no pause for planning, and no break between Sprints. Everything that happens in Scrum happens inside Sprints: Sprint Planning, Daily Scrums, the Sprint Review, and the Sprint Retrospective all take place within the Sprint, and the Sprint is the container for all of them. The fixed length is what creates the rhythm that makes inspection and adaptation reliable. When Sprints vary in length or have gaps between them, the cadence that Scrum depends on breaks down, and forecasting becomes guesswork.

Four rules govern what happens during a Sprint, and the exam tests them directly. First, no changes are made that would endanger the Sprint Goal. The goal is the Sprint's anchor; work can flex, but nothing should be done that puts the goal at risk. Second, quality does not decrease. A team under pressure may be tempted to cut corners to finish more items, but the Definition of Done is not negotiable within a Sprint — reducing quality to hit a scope target is a violation of the framework, not a reasonable trade-off. Third, the Product Backlog is refined as needed during the Sprint, so that future items become clear enough to select. Fourth, scope may be clarified and renegotiated with the Product Owner as more is learned. This last rule is the safety valve: when the Developers discover the work is different from what was expected, they talk to the Product Owner and adjust the scope of the Sprint — while protecting the Sprint Goal and the quality standard.

Each Sprint produces an Increment: a concrete stepping stone toward the Product Goal. The Increment must be usable and must meet the Definition of Done. Multiple Increments may be created within a Sprint, and the sum of the Increments is presented at the Sprint Review. The Sprint is also the unit of forecasting: by observing how much a team completes in a Sprint, and how that changes over time, the Product Owner can forecast progress toward the Product Goal. This is empiricism applied to planning — forecasts based on observed delivery, not on upfront promises. The exam sometimes asks about the relationship between Sprint length and risk: shorter Sprints reduce risk and increase feedback, because the team inspects and adapts more often, but each Sprint still carries the overhead of its events, so the right length depends on the product and the environment.

Cancelling a Sprint

A Sprint may be cancelled, but only in one circumstance and by one person: if the Sprint Goal becomes obsolete, and only the Product Owner has the authority to cancel it. This is a deliberately narrow rule. A Sprint is not cancelled because the team is behind, because a stakeholder changed their mind about a feature, or because the Sprint is going badly — those are situations to inspect and adapt within the Sprint, not reasons to abandon it. Cancellation happens when the goal itself no longer makes sense, for example if the company changes direction or the market shifts so fundamentally that the Sprint's objective is pointless. The consequences are significant: Sprint cancellations consume resources, are traumatic to the Scrum Team, and are very uncommon. When a Sprint is cancelled, completed and usable work is re-evaluated, and any incomplete Product Backlog items are re-estimated and put back into the Product Backlog. The exam tests cancellation frequently, and the correct answer almost always turns on two facts: only the Product Owner can cancel, and only when the Sprint Goal is obsolete.

Worked scenario

Midway through a two-week Sprint, the company's largest customer cancels the contract that the Sprint's work was for. The Sprint Goal — delivering the integration that customer needed — is now meaningless. The Product Owner cancels the Sprint. The completed work that is usable is kept and re-evaluated; the incomplete items go back to the Product Backlog for re-estimation. The team holds a Sprint Retrospective to inspect what happened, and a new Sprint begins with a new goal. Contrast this with a different situation: the team is simply behind schedule because an item was bigger than expected. In that case nobody cancels anything. The Developers and the Product Owner renegotiate scope, the Sprint Goal is protected, quality is maintained, and the Sprint continues. Knowing which situation calls for cancellation and which calls for adaptation is exactly the judgement the exam tests.

Shorter or longer Sprints?

Choosing a Sprint length is a trade-off. Shorter Sprints — one or two weeks — give faster feedback, lower risk, and more frequent adaptation, because the team inspects and adapts more often. They also carry more event overhead relative to the work: Sprint Planning, Review, and Retrospective happen more frequently. Longer Sprints — three or four weeks — give the team more time to complete complex work and reduce the relative overhead, but feedback comes more slowly and the cost of going in the wrong direction is higher. There is no universally right answer; the Guide only sets the maximum at one month and requires consistency. What matters is that the length fits the product and the environment, and that it stays consistent so the team can forecast from observed delivery. The exam occasionally asks about this trade-off, and the correct answer is the one that connects shorter Sprints with faster feedback and lower risk.

The Sprint is also the unit of forecasting. By observing how much the team completes in each Sprint, the Product Owner can forecast progress toward the Product Goal — not by guessing, but by looking at what has actually been delivered. A team that consistently completes five items per Sprint can forecast that a backlog of twenty similar items will take about four Sprints, and the forecast improves with every Sprint of observed data. This is empiricism applied to planning, and it is why consistent Sprint length matters: if the length changes every Sprint, the observations are not comparable and the forecast is meaningless. Stakeholders who want certainty should understand that an empirical forecast is honest about its own uncertainty — it gives a range that narrows over time, rather than a precise date that is wrong.

Advertisement

Exam traps to watch for

Practice questions

Test yourself on this guide. Answers are checked instantly, with a full explanation for each question.

Q1.What is the maximum length of a Sprint?

Report an error

Q2.When does a new Sprint start?

Report an error

Q3.During a Sprint, which of the following is TRUE?

Report an error

Q4.A Sprint Goal has become obsolete because the company changed direction. Who has the authority to cancel the Sprint?

Report an error

Q5.Which of the following is a valid reason to cancel a Sprint?

Report an error

Q6.When a Sprint is cancelled, what happens to the incomplete Product Backlog items?

Report an error

Q7.Which statement about Sprint length is TRUE?

Report an error

Q8.A Scrum Team is under pressure to deliver more in the Sprint and considers skipping testing on some items. What does the Scrum Guide say?

Report an error

Q9.Which of the following events take place WITHIN a Sprint?

Report an error