Scrum for Testers

Scrum is a lightweight framework in which a small team delivers a product in fixed-length cycles called Sprints, each one month or less. The 2020 Scrum Guide defines it through three accountabilities, five events (the Sprint and the four events inside it) and three artifacts, each artifact with a commitment such as the Definition of Done.

In Scrum, a tester is one of the Developers. On a two-week Sprint for a food-delivery app, they do not wait for a testing phase: they question a story before the Sprint, plan its tests in Sprint Planning, test it the day it is built and help decide whether it is done.

Why it matters

Scrum packs planning, building, testing and review into a few weeks. If testing is the last step of the Sprint, untested stories pile up at the end, half-tested work is called done and defects slip into production. The Scrum Guide says quality does not decrease during the Sprint; that holds only if testing is planned from day one.

Knowing Scrum precisely tells you where to act: business-rule questions in refinement, a broken test environment at the Daily Scrum, a recurring quality problem at the Retrospective. And the Definition of Done gives you a quality bar the whole team agreed to.

Accountabilities, artifacts and their commitments

The 2020 Scrum Guide speaks of accountabilities, not job titles. A Scrum Team is one Product Owner, one Scrum Master and Developers, typically 10 or fewer people, with no sub-teams, so there is no separate test team. Its three artifacts each carry a commitment. The Product Backlog, the ordered list of what the product needs, has the Product Goal, the team’s long-term objective. The Sprint Backlog, the items selected for the Sprint and the Developers’ plan for delivering them, has the Sprint Goal, the single objective for the Sprint. The Increment, a usable and thoroughly verified step toward the Product Goal, has the Definition of Done.

Product Backlog refinement, breaking items down and adding detail, order and size, is an ongoing activity rather than an event, and the cheapest place for a tester’s questions. What each accountability means for you:

  • Product Owner: accountable for maximising the product’s value and for managing the Product Backlog, including the Product Goal and the order of items. Questions about business rules and priority go here.
  • Scrum Master: accountable for establishing Scrum and for the team’s effectiveness, including getting impediments removed, such as an unstable test environment.
  • Developers: everyone who builds the Increment, testers included. They own the Sprint Backlog, instil quality by adhering to the Definition of Done and adapt their plan daily toward the Sprint Goal.

What the tester does in each event

The Sprint contains the other four events, lasts one month or less, and the next Sprint starts as soon as it ends. The Daily Scrum is always 15 minutes; the other time-boxes below are maximums for a one-month Sprint, and in shorter Sprints those events are usually shorter.

One Sprint in two lanes: Product Backlog refinement, the four events inside the Sprint with their time-boxes, and the daily work, each paired with what the tester does in it. Conceptual model
  1. Scrum event or activity:
    1Product Backlog refinementOngoing activity, not an eventItems broken down and given detail, order and size
    What the tester does:
    Question and clarifyRules, edge cases, risks: is each item testable?
  2. Scrum event or activity:
    2Sprint PlanningUp to 8 hoursWhy, what and how: the Sprint Goal, the items, the plan
    What the tester does:
    Plan the testingTest tasks, test data, environments and regression scope in the Sprint Backlog
  3. Scrum event or activity:
    3Daily Scrum15 minutes, every working dayDevelopers inspect progress toward the Sprint Goal
    What the tester does:
    Make testing visibleWhat is tested, what is blocked, what is waiting for testing
  4. Scrum event or activity:
    4Work in the SprintEvery day of the SprintDevelopers turn the selected items into Done work
    What the tester does:
    Test continuouslyPair, automate, explore, report defects, run confirmation and regression tests
  5. Scrum event or activity:
    5Sprint ReviewUp to 4 hoursResults presented to key stakeholders; next steps agreed
    What the tester does:
    Share the evidenceWhat was tested, known risks, feedback recorded
  6. Scrum event or activity:
    6Sprint RetrospectiveUp to 3 hours; ends the SprintInspect individuals, interactions, processes, tools and the DoD
    What the tester does:
    Propose an improvementBacked by evidence: escaped defects, waiting time, flaky tests

Throughout steps 1–6 All of this happens inside the Sprint, which lasts one month or less; the next Sprint starts as soon as this one ends.

The Daily Scrum is always 15 minutes; the other time-boxes are maximums for a one-month Sprint, and shorter Sprints usually have shorter events.

  • Sprint Planning (up to 8 hours): the team decides why the Sprint is valuable, what can be Done and how. Put testing into the "how": test tasks, test data, environments and regression scope.
  • Daily Scrum (15 minutes every working day, for the Developers): inspect progress toward the Sprint Goal and adapt the plan. Say what is tested, what is blocked and what is waiting for testing.
  • During the Sprint: test each item as soon as it is ready, not in a batch at the end. Pair with developers, automate checks, explore, report defects, and run confirmation and regression tests.
  • Sprint Review (up to 4 hours): the team presents its results to key stakeholders and they agree what to do next. Share what was tested and the known risks, and make sure feedback reaches the Product Backlog.
  • Sprint Retrospective (up to 3 hours): the team inspects individuals, interactions, processes, tools and its Definition of Done. Bring evidence, such as escaped defects or waiting times, and propose one concrete improvement; it may even go into the next Sprint Backlog.

Try it: Risk-Based Test Generator Try it: Regression Scenario Generator

The Definition of Done and the tester

The Definition of Done (DoD) is a formal description of the state of the Increment when it meets the quality measures required for the product. An item that meets it becomes part of the Increment; an item that does not cannot be released or even presented at the Sprint Review, and returns to the Product Backlog. An organisation’s DoD is every team’s minimum; without one, the team creates its own.

The ISTQB Foundation syllabus v4.0 notes that Agile teams often call their exit criteria the Definition of Done. The tester helps make the DoD checkable, applies it honestly and proposes changes at the Retrospective. A DoD for a bilingual web product might read:

  • Acceptance criteria verified on the test environment.
  • Automated checks for the new behaviour pass in the pipeline.
  • Regression tests for the affected areas passed.
  • No open Blocker or Critical defects on the item.
  • Arabic and English screens checked, including the right-to-left layout.

Try it: Acceptance Criteria Generator

Worked example: one story through a two-week Sprint

A food-delivery team runs two-week Sprints with the Sprint Goal "Returning customers can reorder a past order in two taps". It is a story like the reorder story in the Agile for Testers exercise; this time, follow where each part of the tester’s work happens in Scrum:

  • Refinement, in the previous Sprint: the tester asks what happens if an item from the old order is out of stock or its price has changed. Two acceptance criteria are added (the basket uses today’s prices, and the customer is told which prices changed), and out-of-stock handling becomes its own item.
  • Sprint Planning: API checks for the reorder request, test data with a changed price, an exploratory session on the basket and checkout regression join the plan.
  • Days 4 to 9: the tester raises missing test data at a Daily Scrum and a developer helps after the event. A defect (the old price is copied) is fixed, confirmation-tested and given an automated check the same day; on day 9 the item meets the DoD.
  • Review and Retrospective: a stakeholder’s new request goes to the Product Backlog; the tester shows that three items waited two days for test data, so the team now identifies test data in Sprint Planning.

Try it: User Story → Test Cases Try it: User Story → Test Scenarios

Common mistakes

  • Running the Sprint as a mini-waterfall: build first, test in the last days, carry untested items over.
  • Calling an item done because the code is merged while DoD checks are still open.
  • Using the Daily Scrum for status reports or long defect discussions instead of inspecting progress toward the Sprint Goal.
  • Confusing the Definition of Done with acceptance criteria, or keeping a private checklist instead of improving the team’s DoD.
  • Treating quality as the tester’s job alone; in Scrum every Developer is accountable for it through the DoD.

The QA perspective

On a Scrum team you are a Developer with a testing speciality: agree acceptance criteria and risks with the Product Owner, testability and automation with developers, and impediments with the Scrum Master. Record test notes and defects on the backlog item itself so its real state is visible to everyone.

Every Sprint, ask: can this item be built and tested inside the Sprint, what data and environment does it need, and what could it break? When an item misses the DoD on the last day, say so: it returns to the Product Backlog.

Practice exercise

Plan your testing across one Sprint

Your task

A clinic booking web app runs one-week Sprints. Story: "As a patient, I want to cancel my appointment online so that I do not have to call the clinic." Acceptance criteria: an upcoming appointment can be cancelled from "My appointments", and not less than 24 hours before it.

The team’s DoD: acceptance criteria verified; automated checks passing; no open Blocker or Critical defects; Arabic and English pages checked. Write what you would do as the tester in refinement and in each event, then propose one improvement to the DoD.

Steps
  1. Refinement: write the questions the acceptance criteria leave open.
  2. Sprint Planning: list the test tasks, test data and regression scope you would add to the plan.
  3. Daily Scrum, Sprint Review and Sprint Retrospective: write one thing you would do or say in each.
  4. Compare the DoD with what this story could break, and propose the one improvement that matters most.
Show the model answer
Model answer

Refinement: Is the 24-hour limit counted in the clinic’s time zone, and is cancelling exactly 24 hours before allowed? Does the freed slot become bookable again? Who is notified? The answers become acceptance criteria or new items.

Sprint Planning: boundary tests just under, exactly at and just over 24 hours; API checks for the cancel request; matching test data; an exploratory session on "My appointments"; regression of booking and rescheduling. Raise the unspecified time-zone rule as a risk.

During the Sprint: test the flow the day it reaches the test environment, and report progress and blockers, such as missing test data, at the Daily Scrum. At the Sprint Review, share what was tested and any risk still open, such as the time-zone rule; the item is presented only if it meets the DoD. At the Retrospective, bring evidence such as waiting times or escaped defects.

DoD improvement: add "regression tests for the affected areas passed", because the current DoD does not protect existing booking features.

Frequently asked questions

Is there a tester role in Scrum?

Not as a separate accountability. The 2020 Scrum Guide has three: Product Owner, Scrum Master and Developers, and the Developers are all the people who create the Increment. A tester is one of the Developers. Many teams still have testing specialists, who lead test analysis, exploratory testing and automation while the whole team stays accountable for quality.

What is the difference between the Definition of Done and acceptance criteria?

The Definition of Done is one set of quality measures that applies to every item and to the Increment as a whole. Acceptance criteria are conditions for one specific item. An item is done only when it meets its acceptance criteria and the Definition of Done.

What happens to a story that is not done at the end of the Sprint?

It is not part of the Increment, so it cannot be released or presented at the Sprint Review. It returns to the Product Backlog, where the Product Owner orders it again. The Sprint is not extended to finish it.

Does Scrum require user stories or a Definition of Ready?

No. The Scrum Guide does not prescribe user stories, acceptance criteria or a Definition of Ready; teams add them as practices. The guide only says that items that can be Done within one Sprint are ready for selection in Sprint Planning. The ISTQB Foundation syllabus v4.0 describes the Definition of Ready as the entry criteria for a user story.

References

Practise with Mutqan tools

Convert a user story (As a / I want / So that) into structured test cases using templates and heuristics.

Test Design

Score features by likelihood and impact, then generate a prioritised risk-based test plan.

Test Design

Go deeper

Your progress on this topic

Part of the QA Roadmap