Wednesday, April 24, 2013

Developing a Sprint Goal

The sprint goal is used to help focus the Scrum Team on the objectives of the sprint, the higher purpose of why the sprint is necessary. The Scrum Guide has this to say about creating a sprint goal:

"After the Development Team forecasts the Product Backlog items it will deliver in the Sprint, the Scrum Team crafts a Sprint Goal. The Sprint Goal is an objective that will be met within the Sprint through the implementation of the Product Backlog, and it provides guidance to the Development Team on why it is building the Increment.

By the end of the Sprint Planning Meeting, the Development Team should be able to explain to the Product Owner and Scrum Master how it intends to work as a self-organizing team to accomplish the Sprint Goal and create the anticipated Increment."

My team's Sprint Goal was, at times, essentially a summation of the user stories but the better Sprint Goals in my mind were those that served as a milestone for the development team, the customers, the business, or any combination of these. For example a Sprint Goal for the first sprint of a newly formed Scrum Team might be, "Show that we can add value to the customer using scrum." To achieve this goal the team would need to complete at least one story. You'll note that this Sprint Goal has two parts: one part is about adding value for the customer and the second part is about the Scrum Team realizing their own potential to deliver value using scrum.

In order to keep the development team motivated throughout the sprint,  I would suggest helping the team find a higher purpose and develop a metaphor to describe that purpose. Look for the benefit or who benefits from the sprint and coach the team to use this information in crafting their Sprint Goal.

During the daily scrum, I would ask the development team if any impediments might put achieving the Sprint Goal at risk. I would discontinue this once the development team themselves began alluding to the Sprint Goal in the daily scrum as they planned the day’s activities. My intent was to re-enforce the Sprint Goal as the primary focus rather than the stories, which served as a means to achieving the Sprint Goal.

The Sprint Goal and Shorter Sprints

If the sprint length is short, one or two weeks, then the Sprint Goal might not have the bite or grab the imagination as firmly as when sprints are longer. It could be that by the time the Scrum Team fully embraces the Sprint Goal, the sprint is over. The probable issue here is two-fold: 1), the development team, being technical in nature, will tend to associate themselves with the work rather than the reasons for that work, and 2), the Scrum Team may not be spending a lot of time developing the Sprint Goal for shorter sprints thinking time is at a premium.

As scrum master, you can encourage the Scrum Team to think about the bigger picture and understand the benefits and who benefits from the forthcoming sprint. The length of the sprint fades into the background as the team concentrates on these benefits. The amount of time taken to develop the Sprint Goal is rewarded by the development team gaining insight to the customer's, end user's, or business's needs, and satisfying these needs or moving toward the satisfaction of these needs is the real intent of the sprint.

Technical Sprint Goal

When the Scrum Team first formed up, Sprint Goals tended to be a reflection of the specific stories in the sprint, that is, they tended to focus on the technical nature of the sprint rather than the beneficiaries. This is most probably due to the former way software development was run using waterfall techniques. In a waterfall environment, the development team was not always privy to the reasons behind the requirements and the product owner, often a former BA, product manager, or project manager, hasn't yet learned the benefits of having development team understand the driving force or purpose behind the requirements.

In this scenario the product is the principal beneficiary of the effort rather than any customer, user, or business. The scrum master needs to show the product owner, and quite probably the business, that having the development team involved with the customers, end users, and the business is necessary to getting a better product, increasing the number of customers, and growing the business. The development team needs to understand the customer, end user, and business so better for them to put themselves in their position and visualize the customers', end users', and business's point of view.

Unfocused Sprint

Sometimes the product owner has prioritized the product backlog to do several features or functions at the same time rather than concentrating on one at a time. This probably is another throw-back to the waterfall days when all developers were kept busy on several aspects of the project at the same time. The result is that everything is "in progress" but nothing is "done".
There are a couple of reasons that this approach can be considered flawed. The first is this kind of approach can be construed as going against the agile principle, "Our highest priority is to satisfy the customer through early and continuous delivery of valuable software." By working several features at the same time over several sprints, the team runs the risk of not delivering value at the end of each sprint. A customer may not be particularly thrilled being at sprint review after sprint review without seeing a completed feature.
The second potential flaw is the lack of flexibility to release early with only a partial set of planned functionality completed. For example, if a planned release has 5 new features, it's entirely possible for the business to release the product with only 3 of these completed if it can be determined that the product in its current state achieves "product/market fit." In other words, the cost of completing the remaining 2 features is greater than the perceived value of the product if these 2 were completed. If the product owner elects to do all 5 features in parallel, then the opportunity to release early, and begin generating revenue sooner, is lost.

Not Meeting the Sprint Goal

The Scrum Guide states that if the Sprint Goal becomes "obsolete", then the product owner may cancel the sprint. Since this implies a sudden, massive shift away from the perceived business value of the work in the sprint, this will be a very rare occurrence.
It's entirely possible for a development team to not meet the Sprint Goal but this doesn't mean the team has failed or that the sprint should be cancelled. What this most likely means is an unknown factor has made itself apparent during the sprint causing a slow-down of progress or a change in direction regarding design or technology. Not meeting the Sprint Goal is something that happens to all Scrum Teams and shouldn't be taken as a personal team failure but rather as a call to understand the root causes and correct them. Usually missing a Sprint Goal is due to an insufficient understanding of the work pulled into the sprint or, with less experienced Scrum Teams, over-committing and taking on more work than the team can handle.


Select a Sprint Goal that focuses on the beneficiaries of the sprint using the work as a guide. As a team, try to develop a metaphor to describe what the objective of the sprint is. Avoid the temptation to develop a Sprint Goal that reflects the work, i.e. product, over the beneficiaries.

With shorter sprints the Sprint Goal may not have enough impact on the team. The team should spend the time necessary to develop a Sprint Goal that adds meaning to the sprint even if the sprint is one week long. All sprints, regardless of length, should add value to the customer, end user, or business and the work in the sprint is a means toward achieving this value.

Try and keep a tight focus on the sprint and Sprint Goal. Working one feature at a time can get value to the customer quicker than working several features at once. It is better to have one feature "done" than to have several features "in progress" at the end of the sprint.

Do not get overly anxious if the Sprint Goal will be or is missed and don't be tempted to abort the sprint because of it. Use the sprint retrospective to analyse the root causes and make the appropriate corrections for the next sprint.

Friday, April 12, 2013

Agile Project Manger or Scrum Master?

I find that there's a tremendous amount of prejudice against certain words in the Scrum community and "manager" is near the top. A business can and will give their roles any title they feel is appropriate. I would bet the farm that in some companies, the Scrum Master role is more aligned to a Project Manager of old and that Agile Project Manager role is more aligned to the Scrum Guide’s definition of the Scrum Master. So far no help, right?

Most businesses that are or will be transitioning to Scrum will have an Agile champion on board to guide the transition. I was this person at my previous company and after some time the role of Scrum Master was identified and added to HR’s list. However, before that happened, the person in the Scrum Team coaching and helping the developers and product owner was known as “Scrum Master”. This happened due to training and coaching the software and product management departments received during the transition to Scrum. The internal view was that the servant-leader of a development team was called “Scrum Master”. The external view of the role as projected by the HR department, possibly following 10 – 20 years of tradition, might have still have the role listed as “Project Manager” but now with the added term “Agile” plopped in front. This shouldn’t be too much of a surprise as the adoption of Scrum starts within the software department and will only spread once the successes due to the application of Scrum are recognised. As the Agile champion, my focus was on the software department and not HR, (in the beginning). After a couple months and Scrum had an obvious foothold in the company, efforts to re-align HR roles and role descriptions (Scrum Master, Development Team Member, Product Owner) began.
As a practicing Scrum Master and I saw a job ad that said, “Agile Project Manager”, I would apply with the knowledge that the role is most likely that of a Scrum Master for a company in transition (or possibly stuck in transition). I might find out otherwise at the interview but that’s the point of the interview: the prospective employer and prospective employee determine if there exists a common understanding of expectations and responsibilities surrounding the role. Some may see this as waste but as the interviewee, I see interviews as a learning experience even when the ultimate outcome is no job.

One of the responsibilities of the Scrum Master in the Scrum Guide is, “Leading and coaching the organization in its Scrum adoption”. This is where it can be argued that the Scrum Master is charged with aligning the roles and role descriptions in HR with the roles and descriptions found in the Scrum Guide. If the Scrum Master(s) are unsuccessful in doing this today, then it means they try again tomorrow, and the next, and so on. As long as the internal application of Scrum roles is correctly aligned with the Scrum Guide, then this view external of the Scrum Team might be an annoyance but not necessarily a terminal problem.


In the final analysis the Agile coach and Scrum Masters need to pick their battles with an eye on impact and results. Getting management and the Scrum Team following the Scrum framework is to me, job one. Pulling the rest of the corporate structure in line with Scrum role descriptions i.e., HR and management, is of secondary concern in my view.

Friday, April 5, 2013

How To Recognize A Self-Organizing Team During Sprint Planning

As a Scrum Master, one of  the ways you serve the development team is by coaching them in self-organization.  Several [hundred] definitions of self-organizing exist on the web but I'll use the Scrum Guide and add a point or two.

"Self-organizing teams choose how best to accomplish their work, rather than being directed by others outside the team. ... No one (not even the Scrum Master) tells the Development Team how to turn Product Backlog into Increments of potentially releasable functionality." - Schwaber, Ken and Sutherland, Jeff. "The Scrum Guide", October 2011.

The Scrum Guide definition focuses on the development team understanding and undertaking the work they do during a sprint. The Scrum Guide specifically wants the development team to control how they do the work of the sprint. I would add to that definition two points: the development team acts as their own technical (not line) manager and the development team displays strong professional bearing, unity, and are respectful of others. This manifests itself with the development team understanding and anticipating the technical actions required to implement scrum in the most efficient and expedient manner and the development team working very closely with the product owner (PO) throughout the project.

Self-Organizing During the Sprint Planning Meeting

The Scrum Guide suggests time-boxing the sprint planning meeting at 2-hours for each week of the pending sprint but as the development team becomes more self-organizing, the amount of time necessary for planning should greatly decrease. The two main reasons sprint planning can become shorter are:
  1. The development team has a very good understanding of PBI's coming into the planning meeting primarily due to helping the PO develop the PBI's and conducting product backlog grooming sessions.
  2. The development team and PO work very closely together, discussing customer wants and desires, customer requirements, and customer expectations throughout the project.

Sprint Planning Meeting Part I

The first half of the sprint planning meeting has the PO presenting the ordered product backlog to the development team and the development team forecasting what they can do in the upcoming sprint. In an ideal first half, the self-organizing development team "pulls" off the top x items from the backlog the PO presented and adds these to the upcoming sprint. The development team and PO have already discussed the PBI's during previous grooming sessions but the PO may have made some changes since. The development team and PO review the selected items and any changes made by the PO since the last grooming session. The development team and PO then conjure up a sprint goal and the first half of the sprint planning meeting is done, probably in less than 20-minutes.

Sprint Planning Meeting Part II

The second half of the sprint planning meeting has the development team figuring out how to deliver the PBI's for the sprint. The self-organizing development team will identify enough tasks for the first few days of the sprint. The development team may re-negotiate with the PO if they determine that they cannot do all the PBI's in their present form (i.e. a need to reduce scope) or the development team may find they can do more PBI's and "pull" in these from the product backlog. The second half of the sprint planning meeting now concludes and the sprint starts.
The elapse time for the sprint planning meeting should be less than an hour, under 30-minutes for a two week sprint, when the development team takes on the responsibility to groom the product backlog and understand PBI's to the extent that these PBI's are "sprint ready" i.e., ready to be worked with only minimum review and recap.

A Clue That The Development Team Isn't Fully Self-Organizing

During the sprint planning meeting, the product owner pushes the requirements onto a "dazed" development team.
Picture a sprint planning meeting with a team only recently introduced to agile and scrum. In the first half of this meeting, the PO is presenting product backlog items to the development team. A development team member asks why a story [feature, requirement] is in the upcoming sprint. The PO and development team spend considerable time discussing the necessity and merit of each PBI for the upcoming sprint. In this scenario, there's no indication the development team understands the product backlog items and are merely reacting to what the PO has presented.
There are a few reasons this kind of behavior might be present:
  1. The development team is in command & control mode and doesn't feel it's their place to "pull" work,
  2. the product owner is in command & control mode and feels the need to "manage" the development team,
  3. the development team is not grooming the product backlog with the product owner, or
  4. the development team doesn't understand the product backlog is ordered.
It can be the hard for a development team used to command & control to become self-organizing, especially if the team is transitioning from waterfall. Similarly, if the PO is formally a product manager or project manager, they may have difficulties letting go of the command & control aspects of their former roles. Either of these two is likely to occur and have much the same solution, the scrum master needs to train these people on the philosophy of agile, specifically agile principles:
  • Business people and developers must work together daily throughout the project.
  • The best architectures, requirements, and designs emerge from self-organizing Teams.
  • Build projects around motivated individuals. Give them the environment and support they need, and trust them to get the job done.
These agile principles coupled with the rules and practices of scrum outlined in the Scrum Guide, can start the journey towards a self-organizing team.
The third reason the development team is not involved may be the team isn't working with the product owner to groom the product backlog. Product backlog grooming helps to develop an understanding of the requirements both technically and from a business value (customer) point of view. The scrum master should ensure the development team and PO do grooming and may need to facilitate the initial grooming sessions to make sure this happens. The scrum master can encourage the PO and development team to work together developing PBI's for the customer requirements, develop the acceptance criteria for the customer requirements, and to discuss any risks or issues associated with the customer requirements.
The forth reason a development team isn't "pulling" in work from the product backlog could be they don't quite understand how the product backlog is structured. The Scrum Guide states that the product backlog has the attributes description, order, and estimate and "is often ordered by value, risk, priority, and necessity." Although the PO can decide how to order the product backlog, the product backlog is not the private domain of the PO. There should be no secrets or surprises in the product backlog. The scrum master can help maintain the transparency of the product backlog by facilitating product backlog grooming sessions and by encouraging the PO and development team to break down and write the more detailed customer requirements together.


One of the ways the scrum master serves the development team is to coach them in self-organizing behavior. The development team will become more self-organizing as time passes and one way to see this progress is by observing the way the sprint planning meeting runs. As the development team becomes more self-organizing, one effect should be the swiftness of the sprint planning meeting. This will be primarily due to the PO and development team working in close harmony together as customer requirements are developed, broken down with details added, and the PO and development team sharing a common understanding of the value of each "sprint ready" requirement.