Skip to content
  • Official
  • Beginner
  • English
  • 10 steps · 2 quizzes

Writing User Stories: A Hands-On Workshop with Examples

User story writing workshop with examples: build stories from the template, check them with INVEST, write acceptance criteria and split big stories.

  • Free
  • No account needed
  • Stop any time, pick up where you left off
Workshopy.io

WorkshopyOfficial

78 lessons · used by 2 teachers

For teachers

Teaching this? Run it live with your class.

  • Students join with a code — no accounts
  • See who’s done and who’s stuck, step by step
  • It’s your copy — change anything before class

Sign in to copy it to your workspace. It’s free.

Your path

10 steps · 2 quizzes · open any step, or read it here

  1. Writing user stories: how this workshop worksStart hereIn this workshop you practise writing user stories with examples from real work: the "As a / I want / so that" template, the INVEST checklist, acceptance criteria and splitting a big story.InteractiveHands-on
    Read this step

    In this workshop you practise writing user stories with examples from real work: the "As a / I want / so that" template, the INVEST checklist, acceptance criteria and splitting a big story. Every step has an exercise where you build, check or split a story and see at once whether it holds up.

    A user story is a short description of something a person wants to do with a product, written from their point of view. Teams use stories to plan work in small pieces that each give someone something useful. Press the button below to watch a vague request become a story, and then a set of small, checkable slices.

    Interactive · From a vague request to small, checkable slices — try it in the lesson

    In this lesson you'll:

    • turn a request into a story with a real role, goal and benefit;
    • test a story against the six INVEST qualities and fix the one it breaks;
    • write acceptance criteria and Given/When/Then scenarios a tester could run;
    • split a story that is too big into thin slices that each work on their own.

    It takes about 70 minutes. The examples come from three products: a support team's shared inbox, an expense-claims app and an HR portal.

  2. Card, Conversation, Confirmation: what a story really isA user story is not a specification. Kent Beck introduced stories in Extreme Programming in the late 1990s, and in 2001 Ron Jeffries described their three parts as the 3 Cs: Card: the story written down, short enough for an index card.InteractiveHands-on
    Read this step

    A user story is not a specification. Kent Beck introduced stories in Extreme Programming in the late 1990s, and in 2001 Ron Jeffries described their three parts as the 3 Cs:

    • Card: the story written down, short enough for an index card. It reminds everyone what to talk about.
    • Conversation: the talk between the people who need it and the people who build it. This is where the details come from.
    • Confirmation: the tests that show the story works. Agreed before building starts, they say when the story is done.

    The card is small on purpose. If you try to fit every detail onto it, nobody has the conversation, and the team builds what the card says instead of what the user needs.

    Try it: the notes below all belong to one support inbox story. Sort each one into Card, Conversation or Confirmation.

    Interactive · Sort the notes: Card, Conversation or Confirmation — try it in the lesson

    Takeaway: the written story is the smallest part. Most of the value is in the conversation and the agreed checks.

  3. The template: As a, I want to, so thatThe most common way to write the card is: As a [role], I want to [goal], so that [benefit].InteractiveHands-on
    Read this step

    The most common way to write the card is:

    As a [role], I want to [goal], so that [benefit].

    The format came from a team at the London company Connextra around 2001, and it spread because each part answers a question the team needs answered:

    • Role (who): a real kind of person, such as "a support agent" or "an employee who travels for work". Never "the system" or "a user".
    • Goal (what): something that person wants to do, in their words. Not how to build it ("add a database join") and not a whole product ("manage everything").
    • Benefit (why): the reason it matters to them. It must add something the goal doesn't already say. The benefit is what lets the team suggest a better way to reach it.

    Try it: for each product, read what the team heard and build the story from the parts. A wrong part tells you what is wrong with it.

    Interactive · Build a user story from its parts — try it in the lesson

    Takeaway: who, what and why. If the "so that" just repeats the goal, you don't yet know why anyone wants it.

  4. INVEST: six checks for a good storyIn 2003 Bill Wake proposed INVEST, a checklist for a good story. Mike Cohn's book User Stories Applied (2004) spread it widely. A good story isInteractiveHands-on
    Read this step

    In 2003 Bill Wake proposed INVEST, a checklist for a good story. Mike Cohn's book User Stories Applied (2004) spread it widely. A good story is:

    Letter Quality The question to ask
    I Independent Can it be built in any order?
    N Negotiable Is there room to agree the details together?
    V Valuable Does a user or customer get something from it?
    E Estimable Can the team say roughly how big it is?
    S Small Does it fit comfortably in one sprint?
    T Testable Can we check when it is done?

    INVEST is a checklist, not a score. A story that fails one letter usually tells you what to do next: merge it, split it, ask a question, or write the test.

    Try it: go through the six letters. Each one shows an expense-claims story that breaks it and the same need rewritten.

    Interactive · The six INVEST qualities, broken and fixed — try it in the lesson

    Takeaway: most broken stories fail on V, S or T: technical tasks dressed as stories, stories that are really epics, and wishes nobody can test.

  5. INVEST checker: find the flawNow check stories the way a team does in backlog refinement. Each story below breaks one INVEST quality more clearly than the others. Find it, then read how a team would fix it. Try it: pick the letter for each of the six stories. Use Hint if you're stuck.InteractiveHands-on
    Read this step

    Now check stories the way a team does in backlog refinement. Each story below breaks one INVEST quality more clearly than the others. Find it, then read how a team would fix it.

    Try it: pick the letter for each of the six stories. Use Hint if you're stuck.

    Interactive · Find which INVEST quality each story breaks — try it in the lesson

    Takeaway: naming the letter turns "this story is bad" into a fix everyone can agree on.

  6. Check: stories, the 3 Cs and INVESTFour quick questions before acceptance criteria.Quiz · 4 questions
    Read the questions

    Four quick questions before acceptance criteria.

    Question 1. What do the 3 Cs of a user story stand for?

    • Customer, Code, Check
    • Card, Conversation, Confirmation
    • Context, Criteria, Commitment
    • Create, Change, Close

    Question 2. In "As a support agent, I want to see a customer's earlier tickets next to a new email, so that I don't ask them to repeat themselves", which part is the benefit?

    • a support agent
    • see a customer's earlier tickets next to a new email
    • I don't ask them to repeat themselves
    • The whole sentence

    Question 3. Which story breaks the V in INVEST?

    • As an employee, I want to see my leave balance, so that I can plan time off.
    • As a developer, I want to upgrade the database server.
    • As a support agent, I want to assign an email to myself, so that two people don't answer it.
    • As a finance officer, I want claims over the limit flagged, so that I check them first.

    Question 4. What does Negotiable mean in INVEST?

    • The price of the story can be discussed with the customer
    • The story can be dropped at any time
    • The details are agreed in conversation, not fixed in the story
    • The team votes on whether to build it
  7. Acceptance criteria: the rules that say when it's doneAcceptance criteria are the Confirmation part, written down. They are short rules the story must meet.InteractiveHands-on
    Read this step

    Acceptance criteria are the Confirmation part, written down. They are short rules the story must meet. A good criterion is:

    • checkable: it passes or fails, with no argument;
    • about behaviour: what a user sees or can do, not how the code works;
    • clear at the edges: "up to 200" and "over 200" leave no gap, so everyone knows what happens at exactly 200.

    Write them before the team starts building, in the conversation. Three to six criteria per story is common. If you need many more, the story is probably too big, and the criteria show you where to split it.

    Try it: the story below has four criteria (the amounts are illustrative). Change the claim and make each rule fire. Look at which rule decides first when two could apply.

    Interactive · Acceptance criteria for an expense claim — try it in the lesson

    Takeaway: each criterion is a test you could run by hand. If you can't imagine running it, rewrite it.

  8. Given, When, Then: criteria as scenariosSome criteria are easier to agree on as an example.InteractiveHands-on
    Read this step

    Some criteria are easier to agree on as an example. Given / When / Then comes from behaviour-driven development (BDD), which Dan North and Chris Matts shaped in the mid-2000s:

    • Given the situation before anything happens (the state);
    • When the one thing the person does (the action);
    • Then what can be seen afterwards (the result).
    Scenario: enough days left
      Given the employee has 10 leave days left this year
      When they request 3 days in March
      Then the request shows as Pending, and their manager is told
    

    Each scenario checks one behaviour. Write the usual case first, then one scenario for each rule that can say no. The most common mistakes are an action hidden in Given, a result in When, and a Then nobody can check ("it works").

    Try it: build the three scenarios for an HR portal story. Each wrong choice says which part it belongs to.

    Interactive · Build Given, When, Then scenarios — try it in the lesson

    Takeaway: a state, one action, a visible result. A scenario a tester can run as written is a scenario that ends arguments.

  9. Splitting a big story: workflow, rule, dataA story too big for one sprint is often called an epic. Don't split it by the layers of the system (database, API, screen): no layer on its own lets anyone do anything.InteractiveHands-on
    Read this step

    A story too big for one sprint is often called an epic. Don't split it by the layers of the system (database, API, screen): no layer on its own lets anyone do anything. Split it vertically, so that each slice cuts through every layer and gives a user something to try. Three ways work for most stories:

    1. By workflow step: request, approve, see the decision, cancel.
    2. By business rule: the simple case first, then one rule at a time.
    3. By data: one kind of data first (annual leave), then the others.

    Other common splits: the happy path before the errors, one user role before the others, a simple version before the faster or prettier one.

    Try it: split the HR portal epic each way, then press "By layer" to see why that split doesn't work.

    Interactive · Split a big story three ways, and one way not to — try it in the lesson

    Takeaway: every slice should be something a user could try and tell you whether it helps.

  10. Slicing practice: keep the vertical slicesHere is a support inbox epic and eight ideas for how to split it. Half are vertical slices. The other half are layers, phases or tasks dressed up as stories.InteractiveHands-on
    Read this step

    Here is a support inbox epic and eight ideas for how to split it. Half are vertical slices. The other half are layers, phases or tasks dressed up as stories.

    Try it: keep the four slices an agent could use on their own, drop the other four, then choose which slice to build first.

    Interactive · Keep the vertical slices of a support inbox epic — try it in the lesson

    Takeaway: build the thinnest slice that works end to end first. The others build on it.

Comments

You’ll sign in with Google before it’s posted.Comments are public — don’t name your students.
0 / 1,500

No comments yet. Be the first to say how it went.

Using it

Workshopy shares this lesson under CC BY 4.0. Anyone may use and change it, with credit to you. Anyone can learn it here, free. A copy in your workspace is yours to change and run live; your students never need an account.

Versions

  1. 1

    Version 1Current

    · First published

Copies are made from the current version. A copy you already have doesn’t change.

More from Workshopy

All 78 lessons
All 14 Business lessons