Blog Archives

Who Should Write UATCs?

Collaboration

Experienced agile practitioners take for granted that detailed requirements are captured as user test cases. For organizations transitioning to agile, this is one of the more challenging practices for them to adopt. Some product owners or business analysts still view the traditional “requirements document” as the way to capture detailed requirements. Because of their attachment to this traditional form of documentation, many resist replacing them entirely with User Acceptance Test Cases (UATCs).  They view UATCs as additional work on top of writing their traditional requirements.

UATCs better convey a sense of how the system is supposed to work by using examples rather than abstract, ambiguous business rules. In a mature agile organization, the end users are best qualified to write the UATCs.  This underscores the product owner’s role: they do not need to gather every detailed requirement – their job is to connect the development team with subject matter experts rather than being a conduit between the two.

Working Software over Comprehensive Documentation

Taken to the extreme, only the highest-level details are written down. This can work for small, simple applications. For large, complex applications, this is insufficient and quickly leads to chaos, disappointment, and frustration. The key is not avoiding writing down requirements, it is capturing them ONCE so that they can be used by product owners, developers, users, and testers and be the one living artifact besides the software.

What options do teams have for adopting the practice of capturing requirements as test cases?

1)      Developers write UATCs – while some might consider this sacrilege, in reality this is a great way to get started with UATCs and establish collaborative bonds between the development team and the end users.  Of course the developers’ initial version of these will usually only reflect what is in the acceptance criteria and what was discussed during grooming, but it provides structure for gathering more details from the users.  Users quickly see the value in this approach because the requirements are written in terms that they can understand – system behaviors. They also see the value in working directly with the developers. Product owners like this because it does not add to their workload and they quickly realize the benefits of the enhanced collaboration. One caveat: developers may need coaching about expressing things as tests and expected results since they may also be accustomed to thinking in terms of catch-all business rules rather than tests.

2)      Developers and users write UATCs together – given some starting point such as a sketch of a web page, the developers and users talk through the expected behaviors for a user story. This feels time-consuming but it directly involves the users in the process of devising the test cases.

3)      Users write UATCs – while this is UATC nirvana, it is important to remember that the developers and users still need to review the UATCs together and in detail. The test cases and end product always benefit from each group’s experience and perspective.

4)      Users and testers write automated UATCs – for applications that are able to leverage automated end user testing tools like Fitnesse, users will likely need assistance with the testing toolset, format, etc.

Reflections:

Be pragmatic about who writes the UATCs. Remember the goal: collaboration that leads to deep understanding of expected system behavior.

Essential Agile Skills

As organizations adopt agile practices, there are several key skills that differentiate high-performing teams from mediocre teams:

1) Working in short timeboxes

For a team that is accustomed to delivering working code every six or eight months, two or four week long iterations can be unfathomable. But doing this is essential to the success of the project in order to get the benefits of continuous code integration into a testing environment and frequent user feedback. The sprint’s timebox forces the business stakeholders to constantly question the scope of each feature at a micro level and decide when to carve out functionality.

2) Splitting work into small pieces

The most common excuse for teams wanting to plan sprints using large user stories is that it is “more efficient” to do all of the coding work together and test it at the very end or in the next sprint. This perceived efficiency is self-deception. Successful agile teams know that they must put working code in front of their users early and often. Remember the old saying that “users often don’t know their requirements until the software fails to meet them?”

3) Communication and transparency

Old-school silos between IT, Product/Marketing, and business operations groups are an agile anti-pattern. Political games, finger pointing, and one-upmanship are a huge waste of an organization’s resources and energy. A grown-up culture of transparency allows all of these players to collaborate and solve difficult challenges. For the Scrum Master, this means posting project information in high-traffic areas of the office. More importantly, it means being open with stakeholders about the challenges that the team or project is encountering. Team members can feel secure speaking freely about impediments and project risks in a public forum without fear of reprisal. The essential skill here is knowing how to voice issues in a way that does not sugar-coat them but does set the stage for a constructive discussion.

4) Swarming, collaboration, team first

“Your side of the boat might be leaking but my side is fine” attitude just doesn’t cut it anymore. Lone gunmen need not apply. The most effective agile team members will gladly work outside of their skill sets and comfort zones if it helps the team succeed. Agile developers welcome and seek opportunities to discuss functionality with end users to understand their goals and needs. They leverage techniques such as User Acceptance Test Cases to capture and understand detailed requirements.

5) Focus and finish mindset

This is another one of those things that is counter-intuitive at first for people new to agile. Coders are used to getting all the coding tasks done so that they can get all of the stuff to the testers. Understanding that high work in process = low throughput is key. Developers working in lockstep with testers results in higher throughput, even if that means that the developers have to swarm on non-development tasks every now and then.

6) Scope negotiation

Scope negotiation means being relentless both when talking about items in the current sprint or when doing release planning. Sometimes it takes two, three, or four conversations about an item to admit that it is not really needed.

7) Talking about requirements at the right level of detail at the right point in time

Skilled agilists recognize that the conversation that happens in a user story writing workshop is at a much higher level than the conversation about a story right before that story is pulled into a sprint. They also shun detailed requirements until the sprint starts, knowing that a large collection of detailed requirements is inventory that is subject to decay.

8) Automated testing

Refactoring is a fact of life as requirements are progressively elaborated. Effective implementation of test-driven development and automated testing blunts the effort for regression testing as the application evolves.

Introducing Scrum to a Team

GoFor a Scrum Master introducing agile to a team, there is a balance between implementing concepts and practices at a pace that does not overwhelm the team and demonstrating the benefits as early as possible to build momentum and support within the organization.

Each situation is different, but let’s assume that an organization wants to use Scrum and there is an experienced Scrum Master who has been asked to lead a team of people that are experienced with waterfall approaches and are new to agile. Also assume a reasonable level of management support. How should the Scrum Master go about implementing agile practices on the team? For an experienced agile practitioner, the practices are second nature so it can be tempting to try to introduce it all at once, but the reality is that it must be gradually introduced in bite-sized chunks. The key question is which practices should be introduced first and which practices must be implemented in the longer term? Here are some suggestions:

Groundwork

Before starting the first sprint, there are some things that need to be put in place:

  1. Secure a Product Owner (PO). In many organizations this might be a product manager or representative from a business unit. This requires a serious time commitment and product insight so it should not be an afterthought.
  2. Secure a team where the people will be dedicated just to this team, rather than juggling several different projects at once. This can be a big shift for organizations accustomed to managing project assignments via resource utilization/optimization models tied to specific skillsets.
  3. Educate the team (including the Product Owner) about agile principles and Scrum. Hiring a Scrum trainer to conduct a two or three day course is ideal, but if lead time or budgets do not allow this, prepare an overview that touches on the agile principles as well as the mechanisms and ceremonies of scrum. The Scrum Reference Card and Build Your Own Scrum  are good tools to facilitate this.
  4. Be careful in comparing agile to waterfall. Disparaging waterfall will not endear agile to people who have delivered successfully using waterfall.
  5. The new PO will require additional training about what the role involves.  If they were a product manager accustomed to working with a project manager (PM), they will need to understand the additional responsibilities that get transferred from the PM to the PO. To get them started, help them understand that they will need to work with the team on a daily basis.
  6. Change the team’s physical workspace to foster improved collaboration.  Sometimes it is as simple as taking down a few cubicle walls. Other times, it involves locating people in the same area if they are spread out around the building. Be thoughtful about this. Done wrong, too many people are jammed into a small area and they can’t move around without bumping into each other. This builds frustration and resentment that leads people to resist and question the approach that brought this change. Also, an area with plenty of wall space is best.
  7. Pick an iteration length. Start short so that the team can get used to the different ceremonies and rapidly learn from its experiences. This also quickly gives the team and the organization a feel for what happens before/during/after an iteration.
  8. Populate the initial product backlog:
    • Conduct a story writing workshop. Utilize a technique such as story mapping to organize the stories. Don’t get hung up on teaching story mapping techniques – just do it by organizing stories that way on the wall. It will actually be very intuitive to people who are accustomed to structured analysis techniques. A big challenge here will be in writing the stories. If a work breakdown structure (WBS) exists as a result of the initial business case or previous project planning that was done before deciding to use agile, it may be possible to use it as a starting point for populating the product backlog.  This depends on the extent to which the WBS aligns with product features vs. technical tasks.  The more aligned it is with features, the easier it is to translate. This is an opportunity to reinforce that product backlog items are features and not tasks. Also, team members may correlate this to the design phase of a waterfall project. The focus needs to be on “what” the requirements are and not “how” they will be implemented. Stories need to be written in just enough detail for sizing, not with detailed requirements.
    •  Get the PO to prioritize the backlog by business value. Don’t be surprised if every item is given a priority of ‘1’ the first time around.
    • Create a Definition of Done. This provides the basis for sizing stories. This should include user acceptance testing (avoid the use of “done” and “done done” at all costs!). This will be challenging for a team accustomed to a UAT phase after long design and development phases. Doing this reinforces that the team should write stories that represent testable feature increments rather than individual technical layers of the application.
    • Conduct a planning poker session where the team sizes an initial set of backlog items using story points. This can be a big change for teams accustomed to having senior team members provide the estimates. During this process, the Scrum Master needs to make sure all voices on the team are heard.  Planning poker is ideal way to do this. The Scrum Master solicits input from all team members, especially when there are significant differences in estimates. So that the team does not feel as though it needs to define every last detail of each story, help team understand that this is not the last time these stories will be discussed; detailed requirements will be covered in depth once the sprint starts.
    • This next step depends on the project and organization.  If the organization views the project as an agile incubator, the team might have the leeway not to commit to an overall release plan for the product until a few sprints have been completed and the velocity is understood. If not, you may need to use an old project planning approach to derive a delivery date to satisfy the organization until a release plan can be formulated. Be very careful about providing enough buffer for unknowns – any dates that are published may become the benchmark against which success or failure of the effort will be measured. If possible, characterize any release dates as preliminary until the team has had a chance to run a few sprints.

The First Sprint

  1. Plan the sprint.
    • Set firm start and end dates for sprint. Go with either a 1 or 2 week sprint. A 3 or 4 week sprint length is not recommended because it is harder to convince PO not to change the contents of a sprint if it is longer. Also, longer sprints involve larger numbers of stories for a larger team, which can be difficult for them to manage. A shorter sprint length allows the team to demonstrate results quickly.
    • Calculate the team’s hours of capacity to do work on user stories during the sprint, factoring in sprint length, time for meetings, vacation time, etc.
    • Conduct the sprint planning meeting. The first part should involve the development team and the PO. Review the user stories on the backlog starting with the highest priority items. The team should ask the PO any questions they need to help them understand how to task out each story. The PO can stay for second part if they want. Create tasks for each story. Make tasks granular enough (max 6-8 hours) so that it will be easy to have multiple people working on each story. Continue to add stories until team’s hours of capacity for the sprint are used up.
    • Ask everyone on the team if they are comfortable committing to delivering these stories in the timeframe of the sprint and if they have any concerns or reservations.  Go around the table and ask each person – it can reveal some insights that people might not otherwise volunteer.
  1. Set up a story/task board in the team area. Even though the team might have software for sprint planning, go with high-visibility, tactile information radiators. This is important because it helps the team learn that it needs to be self-organizing/self-assigning. The team should update the board throughout the sprint, not the Scrum Master. The Scrum Master needs to change culture of having the project plan buried in project planning software like MS Project and only being maintained by the PM. This also starts to build culture of transparency with stakeholders. Invite stakeholders to visit the team’s work area to view status of current iteration. Of course, be careful that they don’t interfere by trying to change priorities or introduce new stories.
  2. Start the first sprint.
  3. Do daily standups. Start by helping team understand the purpose and the basics.
  4. Some people on the team may have heard the agile myth about “no documentation”. This is the time to start establishing culture of lean, maintainable documentation that has value through the lifetime of the product. A new team will only be able to bite off so much new stuff at one time. If they are accustomed to some type of UX or use case document, use that artifact (during the first sprint) as a transition into user acceptance test cases.
  5. Coach team members that the first thing that they need to do when they pick up a new user story to work on is to get with the PO and applicable users to discuss detailed requirements (since the team didn’t hammer out every last detail of the story during grooming or planning, right?). Reinforce with the PO that they need to be prepared to work with team members on a daily basis.
  6. While the Scrum Master’s eventual goal is a team that limits WIP and individuals self-assign tasks, don’t necessarily expect that to happen in the first sprint. Remember – people on the team are probably accustomed to having their tasks assigned to them by the PM or tech lead via project plans.  Initially, the Scrum Master may need to ask for volunteers for tasks, or as team members report that they are wrapping up a task, the Scrum Master may need to ask them which task they will work on next. Team member will typically look at the next available story that no one is working on. Steer them instead to first look for tasks they can work on for stories that are already underway but not yet complete.
  7. Start tracking the team’s burn down and post in the team work area.
  8. Run interference if the PO or other business stakeholders try to change the stories in the sprint. This is another reason to keep sprints short (max 2 weeks) at first. It is easier to argue against changing the contents of a sprint when the next sprint is less than 2 week away.
  9. Conclude the sprint on the planned end date.  DO NOT extend the sprint under any circumstances, even if it will take ‘just one more day’ to finish all of the stories. It is important to establish this early as it speaks to the sprint commitment being firm and the importance of sizing/organizing the work so that it can be completed within the iteration.
  10. Do the retrospective.  Keep the format simple initially. “What went well, what did not go well, ideas for improvement” is an easy format. Do not allow anyone from outside the team to attend. Make sure the PO attends because they are part of the team and this meeting is an important way to build trust between the PO and the developers. Ensure attendees know that the Vegas rule applies. At the conclusion of the retrospective ask attendees if it is okay to share ideas that were generated with outsiders – stakeholders will naturally be curious.  Ideas that the team does not want to share are not shared outside the team.
  11. In preparation for the sprint review, prepare a sprint summary document.  Show the team’s burn down, stories committed and status of each story at end of sprint.
  12. Conduct the sprint review. Encourage as many people from around the organization to attend as possible. This is about
    • Demonstrating results
    • Team members taking pride in their work by showing it off
    • Showing the organization that the team is transparent

The development team members (developers, QA people, etc.) should conduct the demos.

The mechanics of the first sprint described above are relatively straightforward, but the reality is that this is all occurring against the backdrop of organizational change:

  • Team members are experiencing dramatic change.  Their physical work area has changed. They are being asked to commit to delivering something without detailed specifications. No one is telling them which tasks to work on.
  • The relationship between business and IT may be very different when using an agile approach. If the PO comes from a business unit and because of the organization’s culture there exists a lack of trust or poor communication between the business unit and IT, the lack of trust and communication issues will not necessarily disappear overnight.
  • The organization’s culture may not value or emphasize teamwork. Part of management and the Scrum Master’s jobs is to build a culture where teamwork is highly valued.

The Second Sprint

In the second sprint, the Scrum Master continues to emphasize agile principles and the mechanisms of Scrum.

  1. The Scrum Master continues to coach regarding self-assignment of tasks. As the Scrum Master backs away a little bit and the team starts to self-organize, others (such as the PO) may feel that the team needs to be directed and step in to direct the team and tell them what to do.  The Scrum Master must also coach them not to direct the team.
  2. The Scrum Master continues to encourage team members to swarm on stories by default rather than only when necessary. Lessons learned from first sprint (e.g. too many stories in progress at the same time) make the purpose of this apparent.  Swarming will not happen overnight, but the Scrum Master needs to keep encouraging it.
  3. Conduct user story grooming meetings during the sprint. In the meetings, continue the mantra of discussing features at “the right level of detail at the right point in time.” Use these meetings to size out the rest of the backlog and to revisit stories likely to be worked on in the next sprint. The product owner may be unprepared for these meetings because they are already overwhelmed by working with the team on the current sprint’s user stories. Sometimes it helps for the Scrum Master to meet separately with the product owner a day or two before a grooming meeting to survey the backlog and discuss the next set of stories to be groomed.
  4. Encourage the PO to regularly tend to the backlog – not just before grooming or planning meetings.
  5. Using the team’s velocity from the first sprint, begin working with the product owner to formulate the release plan. It may not be appropriate to go public with it yet. Velocity cannot necessarily be trusted after just one sprint, but doing a preliminary release plan is valuable to get the PO thinking in terms of using the velocity and backlog to formulate a plan.
  6. Start tracking the team’s burn up and also post in the team work area.  The burn up emphasizes that hours worked do not provide business value, only completed user stories do. A graph that shows a steady burn up of completed story points during the sprint (rather than a ”hockey stick” where most of the stories get closed at/near the end of the sprint) is the goal. Don’t necessarily beat the team over the head with this…at this point just introduce it to them and call attention to each story as it is closed during the sprint.

Beyond

Over subsequent sprints, the Scrum Master should consider introducing some of the following techniques.  Of course each situation is unique so he/she will have to gauge whether or not the team is ready for each new practice.

  • For teams that are having trouble swarming and have too many stories in progress at the same time, reinforce the concept of WIP limits.  For teams that already have a task/scrum board with steps in the process laid out, the WIP limits can easily be established for each step. The team can decide whether they are firm limits not to be exceeded or soft limits that are triggers for conversations.
  • Help the team learn the finer points of writing user stories. A book study (e.g. “User Stories Applied” by Mike Cohn) is a good way to do this. Be sure to include the PO in this.
  • Test-Driven Development (TDD).  The benefits of this can be difficult for a team to grasp. Once the team (including the Product Owner) understands that it is okay to refactor and that it is simply part of progressive elaboration, the benefits of TDD become much more apparent.
  • Acceptance Test Driven Development, manual or automated.
  • Introduce additional retrospective techniques to help the team learn about itself and identify areas for growth. The Scrum Master will need to judge what type of retrospective is appropriate depending on how the sprint went.

In addition to learning new practices, the team overall needs to learn to be agile, not just to use agile practices.

  • Work with team members’ managers to help them develop a broader, t-shaped skillset that allows them to perform a larger variety of technical and non-technical tasks.  For specialists, this can be a big shift.
  • Help the team to develop an agile mindset by only learning enough about a feature to size and plan it and then really dig into the details of it once it has been pulled into a sprint. One of the most gratifying moments for a Scrum Master can be when the PO says, “We did not think of that aspect of the feature in when we discussed the story. We will just create a new story and handle it in the next sprint” – without taking the team to task for not delivering it in the current sprint.
  • When the Scrum Master has been steering the team towards new practices but the team has resisted using them, the Scrum Master may need to let the team fail for a sprint. The Scrum Master needs to help stakeholders understand that this is part of the learning process, and two-week sprints will minimize most of the damage if the team gets off-track.

Reflections

  • Every situation is different so of course there is no “one size fits all” approach to starting with agile. Organization and team dynamics impact the speed and extent to which agile is adopted.
  • As the team becomes more comfortable with the agile approach, the Scrum Master should adapt his/her approach to be in more of a supporting role for the team.
  • Occasionally the team will forget lessons learned from a few sprints ago and revert to old habits.  The Scrum Master can bring previous retrospective conversations back into focus to help the team avoid making the same mistakes.

Lean Requirements with User Acceptance Test Cases

The Agile Manifesto extols the virtues of working software over documentation as a true measure of progress and business value. Teams in the early stages of agile adoption sometimes misinterpret this to mean that they should not write anything down.  Other teams that are accustomed to waterfall and documentation-heavy methodologies continue creating extensive and redundant documentation under the guise of User Stories and Acceptance Criteria.

Lean Agile Requirements is a middle ground that provides a framework in which complex requirements are captured in living documents that speak to many audiences and serve a purpose beyond the initial development of the application. The flow for discovering and capturing these requirements in the process of developing new software appears in the diagram below.

User Story Writing Workshop

Product Owners, business representatives, and the development team meet to identify user stories. User roles are identified and a couple of sentences are written to describe the goal and value of each story. Stories are mapped into releases and become the basis of the product backlog.

Grooming

Grooming conversations occur at regular intervals in advance of the sprint in which a story is actually turned into software.  In grooming, a user story may just contain a couple of sentences describing the feature increment.  Each story is discussed and the UI might be sketched on a whiteboard or in a UI mockup tool. Acceptance criteria also may be identified – these are high-level user acceptance test cases (hint: these are not detailed tests, yet) but are only discussed/captured to the extent needed to size the story.

Sprint Planning

Sprint planning conversations are where the team starts to think in more detail about how it will build the feature increment described in a story. The discussion becomes more detailed but it is important to avoid getting into too much detail too soon (e.g. avoid things like “Should we show transactions that are less than or less than or equal to the Invoice Date?”). Any details that do surface can be captured as the initial User Acceptance Test Cases but they are not always necessary at this point. The technical design is reflected in the tasks and estimates that the team creates. Note that the user story itself does not need to contain any of this additional detail.

What is a User Acceptance Test Case?

Acceptance Test Driven Development (ATDD) is based on the premise that one of the most effective ways of communicating requirements is by understanding how the user will use/test a feature and the expected results.  Acceptance tests avoid many of the pitfalls of business rule statements because they eliminate the layer of interpretation and ambiguity involved in understanding the outcome of a business rule in various scenarios.  Writing these tests is as simple as asking the user “What tests will you run to validate this feature?”  The test cases capture the user’s actions and expected results using real examples as much as possible. For the developer, it is like having the answers to an exam before taking it.

Sprint

When the team has pulled a user story into a sprint, the first step in working on the story is to meet with the users and detail the user acceptance test cases. This is when and how detailed requirements and behaviors are captured.  Note once again that the user story itself does not need to contain any of this detail. The detailed user acceptance test cases are then used by developers, QA staff, and users to create and validate the working software.

Requirements Life Cycle in Support

Support

Once the feature is deployed to production, bug reports that are handled as support tickets may uncover additional user acceptance test cases. As the one living document that the team maintains, the user acceptance test cases are updated as the software evolves.  They are used to run regression tests to make sure that other functionality is not impacted by the defect correction.

By capturing complex requirements in the form of user acceptance test cases, the team can build and maintain a product with the confidence that the user’s needs are being satisfied.