Cheat Sheet for Backlog Refinement

Backlog Refinement Meeting

What is it?

The purpose of backlog refinement (grooming) is to make improvements to the product backlog.  Though there is no official ceremony detailed in the Scrum Guide, the activity of refining the Backlog is.

Who does it?

Backlog Refinement is a collaborative effort involving:

• (Optional) facilitator – (like a ScrumMaster) keeps the session moving toward its intended goal

• (Optional) representative(s) from the Management Team – clarify the high level priorities

• (Mandatory) representative(s) from the Product Owner Team – clarify the details of the product backlog items

• (Mandatory) representatives from the Agile Delivery Team – define the work and effort necessary to fulfill the completion of agreed upon product backlog items. It is recommended to have at least one developer and one tester when refining the backlog, to ensure alternate viewpoints of the system are present.

When is it?

Before development of a user story in the current sprint (iteration), generally sometime during the previous 1 or 2 sprints, the team sits down to discuss the upcoming work. Do not wait too late to add details, because the delay will slow the team down. Do not refine your stories too far in advance, because the details might get stale. Depending on the delivery rate of your teams, you should be meeting once or twice a week to review the backlog.

Before You Begin

We need to ensure:

• The product backlog is top-ordered to reflect the greatest needs of Management Team and the Product Owner Team

• Candidates for grooming include stories identified as not ready to complete within the next sprint or will require multiple days of research

• Epics, features or other items on the Management Team roadmap are reviewed periodically

The Backlog

The product backlog can address anything deemed valuable by the Product Owner Team. For the purpose of sprint planning, when using Scrum as the delivery framework, product backlog items must be small enough to be completed and accepted during the sprint and can be verified that they were implemented to the satisfaction of the Product Owner team.

Estimate

Backlog items must be completed in a single sprint or split into multiple user stories. While refining, give stories an initial estimate to see if they are small enough. If not, split them. The best way to split product backlog items is by value and not by process.

Acceptance Criteria

All stories require acceptance criteria. Without it, you can not define the boundaries of a user story and confirm when a story is done and working as intended. Ensure acceptance criteria is testable.   This is what your testers should be writing tests against.

Rewrite Written Stories

Ensure the user story format is consistent, INVEST criteria is being met, and is written from a business not technical perspective.  Know who the customer is. It may not be an end user. Rather, the story may be for something like a service, to be consumed by another team.

Image Credit: Pictofigo

Originally Posted at LeadingAgile

A Lack of a Shared Understanding

Earlier this week, I facilitated a backlog refinement meeting.  In the past, the team was used to all of the requirements being completed (in advance) by the analysts. The delivery team would then execute on those requirements. The problem, of course, was no shared understanding. We came into the meeting with everyone agreeing they were on the same page.  That was true for about 15 minutes. The more we talked, the more they realized they were looking at things from individual perspectives.

At the beginning of the meeting, we had less than 10 user stories, from an analyst's perspective. By the end of the meeting, we had a prioritized backlog with over 100 user stories at different levels of granularity.  It's not perfect and it's never done.  But, it's a start.  For the first time, developers and testers were engaged at the beginning.  At LeadingAgile, we call this the Product Owner team.  When the highest priority stories get to a "ready" state, they will be pulled into a delivery team's queue.  Until then, we need to answer some of the more complicated questions, mitigate risk, and achieve that shared understanding.

Image Source: Based on hand drawn image from Pictofigo

New Scrum Team Challenges

A while ago, I published a post titled: Simple Cheat Sheet to Sprint Planning Meeting.  Though I understand every team is different, I thought it would be helpful to those who are new to agile processes.  People are always looking for cheat sheets, templates, and stuff like that.  What is the harm of giving them what they want? In retrospect, I think the harm is the lack of context.  When people come to a training class, they are provided an ideal situation.  Even the Scrum Guide was written as though you've been doing Scrum for a while. It doesn't talk about things that happen leading up to your team's first Sprint.  It doesn't talk about the complexity of scaling to the enterprise.  It's just vanilla.

I just got back from coaching a new team and I'll be heading back next week to see how they are doing.  To get them moving forward, I facilitated release planning and sprint planning.  That's where things started to get interesting.  What if your team has never done release planning or sprint planning?

Release and Sprint Planning

The purpose of Release Planning is so the organization can have a roadmap that helps them reach their goals.  Because a lot of what we know emerges over time, most of what we actually know is that we don't know much.

The purpose of the Sprint Planning is for the entire team to agree to complete a set of ready top-ordered product backlog items. This agreement will define the sprint backlog and is based on the team’s velocity or capacity and the length of the sprint timebox.

You're new to Scrum.  You don't have a velocity (rate of delivery in previous sprints) and you are not sure of your capacity (how much work your team can handle at a sustainable pace).  What do you do?

The Backlog

The product backlog can address just about anything, to include new functionality, bugs, and risks. For the purpose of sprint planning, product backlog items must be small enough to be completed (developed, tested, documented) during the sprint and can be verified that they were implemented to the satisfaction of the Product Owner. 

Again, you're new to Scrum.  How do you know what small enough is?

Right Sizing Backlog Items

Product backlog items determined to be too large to be completed in a sprint, based on historical data of the team, should not be considered as sprint backlog candidates during the sprint planning meeting and should be split into smaller pieces. Remember, each story must be able to stand on its own (a vertical slice).  It should not be incomplete or process based (a horizontal slice).

Again, you're new to Scrum.  You have no historical data.

Determining Velocity

The velocity of a team is derived by summing the estimates of all completed and accepted work from the previous sprint.  By tracking team velocity over time, teams focus less on utilization and more on throughput.

If you have a new team, not only do you not have a velocity, you won't have any completed work to compare your estimates to and you won't know how many deliverables to commit to.  What do you do?  I think you just go for it!  You keep track of what you are completing and collect historical data.  You get through the sprint and establish some context for future sprints.

The first few sprints are going to be pretty rocky, until the team begins to stabilize. Just ask yourself.  WWDD?  What would Deming Do? Plan a little; Do a little; Check what you did versus what you planned to do; Act on what you discover.

PMI Eating its Own Dogfood

I am happy to report that the PMI-ACP Community Guide project is off and running.  Each day, I see new content being added.  I wonder if this is how Jimmy Wales felt in the early days of  Wikipedia.  Our first measure of success is to get the content of each page of The Guide as close to the vision of the ACP Steering Committee as quickly as possible.  Our second measure of success is to reach a tipping point, whereby the community (not the support team) is maintaining the guide.  Remember, future versions of the ACP exam will be based on this Guide.

Community Guide

To reaffirm, there will not be an AgileBOK.  The Community Guide of the PMI-ACP (login required) is an initiative of the PMI Agile Community of Practice to provide ongoing support for the PMI-ACP agile certification.

ACP Support Team

Lead by Joseph Flahiff of Whitewater Projects and myself, the ACP Support Team has kickstarted the Community Guide content creation process.  We are empowered and 100% self-organized.

The Backlog

In order to deliver something of value to the community, Joseph and I leveraged the wiki within PMI's website to create a Product Backlog.  We wanted transparency and for everyone to know what we are focused on.  Every major area of the ACP exam has a page waiting to be edited. If you had a traditional product backlog, the 10 major areas that comprise the Tools & Techniques of the exam could easily be considered Epics.  Each page of our wiki could be compared to a User Story.  We're not estimating our work.  We're just doing it.

Iteration 1

We are currently in Iteration 1, which ends on May 10, 2012.  Of our 15 member team, we asked volunteers to commit to work on the first 7 pages of the first content area.  At the end of each iteration, we can ask members of the ACP steering committee to review what we have done.  It's important that we stay focused, have short feedback loops, and deliver something of value on a regular basis.

Eating the Dogfood

When you think of PMI, you probably think of project plans, schedules, and stuff like that.  As a self-organized and empowered team, we decided what is important, to increase the chances of our success.  Though there should be a predictable date of completion, based on the currently defined deliverables and length of the iterations, we're prepared for things to change.  We may have to rework some of the pages.  We may have some team turnover.  Regardless, we can guarantee we will deliver value on a regular basis.  We can guarantee there is collaboration with the community.

Joining the Team

If you are interested in creating or maintaining articles for the Community Guide, join our team!  If you want to work independently, we welcome your valuable contributions.

Simple Cheat Sheet to Sprint Planning Meeting

WHAT IS SPRINT PLANNING?

Sprint planning is a timeboxed working session that lasts roughly 1 hour for every week of a sprint.  In sprint planning, the entire team agrees to complete a set of product backlog items.  This agreement defines the sprint backlog and is based on the team’s velocity or capacity and the length of the sprint.

WHO DOES IT?

Sprint planning is a collaborative effort involving a ScrumMaster, who facilitates the meeting, a Product Owner, who clarifies the details of the product backlog items and their respective acceptance criteria, and the Entire Agile Team, who define the work and effort necessary to meet their sprint commitment.

HOW DO WE PREPARE?

Ensure all sprint candidates meet the team’s definition of ready.  In the days and weeks leading up to sprint planning, the Product Owner identify the items with the greatest value and works towards getting them to a ready state.

• Assign a relative story point value
• Remove dependencies
• Create testable examples
• Define acceptance criteria
• Meets INVEST criteria

WHAT IS THE BACKLOG?

The product backlog can address just about anything, to include new functionality, bugs, and risks. Product backlog items (PBI’s) must be small enough to complete during a sprint and should be small enough to complete within a few days. All stories must be verified that they are implemented to the satisfaction of the Product Owner. 

ENSURE RIGHT SIZING BACKLOG ITEMS

Based on historical data of the team, first determine if product backlog items are too large to complete in a sprint.  In these cases, do not consider these stories as valid sprint backlog candidates. Rather, in order to consider for sprint planning, split the stories into smaller pieces. Additionally, each story must be able to stand on its own as a vertical slice.  Therefore, stories should not be incomplete or process-based as a horizontal slice.

CALCULATING A COMMITMENT

To calculate a commitment, mature teams may use a combination of both team availability and velocity.  However, new teams may not know their velocity or they may not be stable enough to use velocity as a basis for sprint planning.  In these cases, new teams may need to make forecasts based solely on the their capacity.

DETERMINING VELOCITY

First of all, as velocity is unique to every team, never use another team’s velocity to plan your sprint.  Derive team velocity by summing the story point estimates of all completed and accepted work from the previous sprint.  By tracking team velocity over time, teams will begin to focus less on utilization and consequently more on throughput.

DETERMINING CAPACITY

For teams without a stable velocity, each team member should provide three simple measures to determine capacity.  First, what are the number of ideal hours in their work day?  Second, how many days in the sprint will that person be available?  Third, what percentage of time will that person dedicate to this team?

THE PLANNING STEPS

1. Remind the team of the big picture or goal
2. Discuss any new information that may impact the plan
3. Present the velocity to be used for this release
4. Confirm team capacity
5. Confirm any currently known issues and concerns and record as appropriate
6. Review the definition of DONE and make any appropriate updates based on technology, skill, or team member changes since the last sprint
7. Present proposed product backlog items to consider for the sprint backlog
8. Determine the needs, sign up for work, and estimate the work owned
9. Product Owner answers clarifying questions and elaborates acceptance criteria
10. Confirm any new issues and concerns raised during meeting and record
11. Confirm any assumptions or dependencies discovered during planning and record
12. ScrumMaster calls for a group consensus on the plan
13. Team and Product Owner signal if this is the best plan they can make given what they know right now
14. Get back to work

// If your team is new to Scrum, download a copy of the Sprint Planning Cheat Sheet //

Drawings by Pictofigo

Zero Cost Effect

I had dinner with a colleague the other night.  I inadvertently quoted something verbatim from Dan Pink's book, Drive. My colleague said if I liked Dan Pink's work, I should read something from Dan Ariely.  So, I started on Predictably Irrational:  The Hidden Forces That Shape Our Decisions. Wow, this book is crazy!  I'm not going to go into any more details in the post other than a comparison of an experiment detailed in the book and something I've seen in the real world. In the book, the author described an experiment on 34 Halloween trick-or-treaters. As soon as the children knocked on the door, they received 3 Hershey's (each weighing about 0.16 oz.) and were asked to hold the Hershey’s they had just received in their open hand in front of them. Each child was then offered a choice between a small (1 oz.) and a large (2 oz.) Snickers bar, under a Cost Condition and under a Free Condition.  In the Free Condition, they could simply get the small 1 oz. Snickers bar (for free) without giving up anything or they could exchange 1 of their 3 Hershey's for the 1 large Snickers bar.  In the Cost Condition, the children could exchange 1 of their .16 oz. Hershey's for the small (1 oz.) Snickers bar or exchange 2 Hersheys for the large (2 oz.) Snickers bar.  They could also choose to do nothing but all of the kids chose to make an exchange.

Experiment Results

In the Free Condition, in which the small Snickers bar is free, demand for it increases substantially (relative to the Cost Condition).  The results demonstrate the attractiveness of zero cost.  People gravitate more toward options that do not require giving up anything.

Example of this on a project

At work, I've had a Product Owner (PO) who wanted to add items from the Backlog to the Sprint.  During sprint planning, the team basically added a buffer, to account for unforeseen events.  I know people are going to crucify me for this, but basically, the Product Owner always seemed to want to shift priorities of work mid-Sprint.  Rather than killing the Sprint, we added a buffer.  This would allow new work to be entertained without totally derailing the work already being completed.  Yes, we could have used Kanban and all of this could have been avoided.  But, Kanban wasn't an option.

So, what happened?  I offered the PO a deal.  I could allow him to add a certain amount of work to the Sprint for free. When I did this, he usually asked for smaller deliverables (relative to other items on the backlog that were ready to work).  But, when I said some work would have to come off the table to pay for the new work, he always went big.  He would choose larger deliverables relative to other items on the backlog that were ready to work.

All I can say is we truly are predictably irrational.

---

Yes, the links to the books are affiliate links.

Valentine's Day Pull System

Nothing is quite as romantic as sitting with your husband or wife, sharing a dinner of fondue.  Nothing, that is, unless you’re sitting with me.  I’m not a big fondue guy.  This is sad because my wife loves it.  She enjoys the whole process.  To counter that, I like my food to be given to me, already prepared.  I enjoy the results! So, leave it to me to point at the fondue pot half way through dinner and yell “Fondue is a pull system!  Fondue is a pull system!”

What is a pull system you ask?  Perhaps you’ve heard of Drum-Buffer-Rope or Kanban?

Businessdictionary.com defines it as a Manufacturing system in which production is based on actual demand, and where information flows from market to management in a direction opposite to that in traditional (push) systems.

The idea behind a pull system is to keep a smooth production flow.  For the sake of argument, let’s say N is a volume of work output.  It can be trouble-ticket call volume, software development, or hardware manufacturing.  Any of these work in the example.  If you and your team can consistently deliver quality N in a month, and keep a good work-life balance, you should know that a C-Level executive asking you to deliver N+10 is going to create bottlenecks in your process flow.  Your overall delivery velocity is going to slow and your team is going to work longer hours trying to deliver N+10.  This increase is unsustainable unless something changes.  You need to get back to N, either by cutting back the work, expanding the amount of people or things to complete the work, or find some efficiencies.

You limited your work in progress (WIP) for a reason!  If more is going into your process than is coming out, you’re going to accumulate a backlog.  For every unit exiting your process, you should have another unit ready to enter it.

For fondue, you limit your work in progress by the amount of long-stemmed forks you have to put into the pot (caquelon).  Before you start, you cut up all of your fruits, vegetables, meats...whatever you plan to dip.  That’s your product backlog.  You then begin dipping whatever you have, X at a time.  We had 3 forks each.  When the food is done, you take it off the fork, add another piece of food, and back into the pot it goes.  You can’t eat anything until it’s done with the process.

Am I a romantic buzz kill or what!?

Image: Recipe Tips

Project Management Theater

tsa

After hearing public outcry over all of the "junk" grabbing going on at Transportation Security Agency (TSA) checkpoints, I heard the resurfacing of the term "Security Theater".  I'm not certain if TSA "gets it".  If you are going to take true action to help fix issues, you need to treat the cause and not a symptom.  Have a shoe bomber?  Make everyone take off their shoes.  Have someone wearing baggy clothes wear a bomb onto a plane, spend millions of dollars to see beyond the baggy clothes.  Without telling you what I did, I bypassed both the new full-body scanners and the TSA pat down in two major airports within the last few weeks.  Certainly, I didn't want to deal with either so I was happy.  The problem I have, as a stakeholder, is a lot of money has been spent and the issue still exists.

Imagine if that happened on your project?

I see Lessons Learned meetings or a Retrospectives as opportunities to help you refine your processes.  You see what works and doesn't work.  You find out the root causes and then you make changes.  You refine.

Today I witnessed what I call Project Management Theater.  The vendor loves to use Gantt charts.    On a program level, both the customer and vendor follow a more traditional waterfall process.  At last count (5 minutes ago) the "integrated" schedule had 5,954 lines.  (Internally, I use a backlog and Kanban) Within seconds of reviewing this monster schedule, I could point out improper work decomposition, improper work package mapping, description inconsistencies, improper use of preprocessors or successors and the list goes on.  If your customer prefers the use of Gantt charts over Burndown charts, I'm not going to argue with them.  Whatever the culture will demand, you have to work with it.  But, the problem here is these are just charts.  They are only as good as the data driving them.  When the customer asked me today what I thought of the split view the vendor provided (WBS/Gantt chart), I was blunt.  I hate it.I added, everything that needed to be reviewed at the meeting could have been presented either as a milestone report or backlog.  Instead, we spent most of our time trying to locate activities and get statuses on each.  On top of that, the schedule provided had not been updated in two weeks.  Therefore, we had to ask over and over again if certain activities had been completed.

If you're going to commit time and money for a support activity, please make sure the resulting "thing" has some value.  At the next meeting, I expect the Gantt chart to go the way of the dinosaur.  I'm advising the customer to request a milestone report from the vendor (instead of the WBS/Gantt Chart).  In the end, I want to ensure the vendor is reaching agreed upon milestones.  Currently, the customer is so distracted by all of the inaccurate details of the schedule, they forget to ask the hard questions about the milestones.

Eliminating the Gantt chart is not going to solve the problem.  Next week, I'm going to show the executive team a Kanban of the milestones.  Let's see if they find more value in that.

Like the image? Find it at Pictofigo