Mike Kuphal's Thoughts

Recently, a colleague sent me a link to a Agile Finland Seminar; Jim Coplien: Ten things that Scrum teams almost always do wrong . The talk is about 2 hours long, stuffed with excellent information, even if Coplien’s presentation style is a bit more laid back than I prefer. In it, he references that he does ‘Red Pill Scrum’ as opposed to ‘Blue Pill Scrum’.

Being an avid Matrix movie fan, I couldn’t help but think of Morpheus’s epic speech to Neo regarding if he wanted to know the ‘true’ meaning of the Matrix or not. If you have seen the movie, you know what I am talking about. Neo chose the Red pill… and the reality it lead to was certainly a game changer!

Coplien states Blue Pill Scrum is defined as simply passing the Nokia test (team focus). For many Scrum teams, ‘simply’ passing this test is difficult enough. (By the way, like the article states, many don’t even pass the Iterative Development part of the test!)

Red pill Scrum also integrates management and enterprise usage of Scrum from top to bottom.

Coplien states blue pill will give you 10% – 20% improvement. Red pill yields one to two orders of magnitude performance improvement.

I personally feel that organizations that are just starting to adopt Agile/Scrum need to experience team success to gain momentum and ‘social proof’ for other teams to see. Once that has been successfully done, moving to enterprise Scrum usage makes sense and the cross department synergy that will result will certainly yield crazy good organizational performance improvement.

I highly recommend giving the video seminar a view. Jim Coplien is one of the founders and proponents of Agile software development. He works closely with Ken Schwaber, Jeff Sutherland, David Starr and others to facilitate Scrum’s evolution as formalized in the Scrum Guide.

Once your team has decided to go forward with Group planning and estimating (and may have even decided to try Planning Poker as a estimation technique), often the next question is: “How much time should I expect a planning session to take?”

Best Practice: Time box 1-2 hours of planning per week in Sprint

A common rule of thumb quoted all over the internet is to time box sprint planning to 1-2 hours per week in the sprint.  So expect 2-4 hours for a two week sprint, 4-8 hours for a four week sprint.  In my experience, I have found that this amount of time is more than enough as long as your team has some experience with group planning.  If your team is new to group planning (or a majority of your team is new to it), expect to add an additional hour or two just to allow for learning discussions to take their course.  Expect 3-5 sprints to allow new team members to really get the hang of sprinting and the planning process that goes with it.

To some, 2-4 hours of planning time, an additional hour for a Retrospective, and also an hour for Review/Demo, seems like a lot to allocate for each team member every two weeks (assuming two weeks sprints).  In essence, you have a full day of ‘Meetings’ for the team.  My experience has been that the team can feel that this is a lot, but it really does pay off over the two week sprint, by ensuring all team members have a good understanding of all tasks and has participated in ensuring any unique knowledge a specific team member has about a task is shared and included in the estimation process.  This also allows for the knowledge sharing necessary to allow team members to pick their tasks off the task board as the sprint progresses, allowing for a team ‘to do’ task board instead of a number of personal ‘to do’ queues.

Best Practice: Start your Sprints on Tuesday, Wednesday, or Thursday

One other important thing to consider: Attempt to start your sprint, and therefore your planning sessions, on a day other than the start or end of the week (Mon/Fri).  There is a natural build up to the end of a sprint, and it’s just not fun to end a sprint on a Friday or Monday given the weekend in between.  This also takes into account that often team members will extend their weekend by taking a vacation day on these edge days and might miss a planning meeting more often.  Teams I participate on have started sprints on Tuesday, Wednesday, or Thursday and have found those days have worked well.

Next in this series: Bugs/Defects: How they fit into the process.

In my last post, I talked about the importance of tracking Remaining Work Hours instead of Completed Hours and some of the reasons why this can work for your team.  Because of the relatively short nature of a Sprint (usually 2-4 weeks), one might think it’s not important to update the Remaining Work hours on their individual tasks, but instead just set the task to ‘in progress’ when work starts on them, and when completed, remove the Remaining Work hours and set to done.  Truth be told, many times this would be just fine.  It might make the Sprint burn down chart a bit more ‘up and down’, but the work would get done just fine.

I strongly encourage teams I work with to update the Remaining Work hours at least once a day on any active tasks.  Some have accused me of attempting to ‘Micro-Manage’ the team by this request.  I simply say to them now: “You can micro-manage yourself if you want, this request is for the benefit of the team, not me.”  By updating the Remaining work hours daily, the whole team has a better picture of how all tasks are progressing, and it allows the team as a whole to quickly identify possible time line issues and adjust to the ebbs and flows of all development work.

The one big caveat that goes with asking the team to update their Remaining work hours daily, is the time entry mechanism has to be simple, fast, and readily available. If it takes more than 30 seconds to update this metric, it simply will not get done consistently.

Though I understand why Remaining work hour entry time (tool wise) can be a barrier to getting it entered, I believe the majority of the time Remaining Work hours don’t get updated daily has to do more with the thinking necessary to attempt to truthfully estimate what is left at the time of updating.  Though that can be challenging, it’s worth the effort to checkpoint yourself each day given the new info (or lack there of) you are exposed to.  Also, doing this checkpoint daily will get quicker and quicker as time goes on given the amount of practice you will get doing it.

Next in this series: Best Sprint start days and how long you should expect (on average) a sprint planning session to take.

Teams I Scrum Master do not track completed hours.  Remaining work hours are the ticket.  That is where the real information is held.  This is not an easy transition for the Project Manager in all of us.  Even for the team members, many of whom have been asked to track completed work hours for years and years, it can be difficult to give up the mentality of reducing a bank of hours predefined prior to the task starting. 

So you ask: by tracking completed hours, don’t you simply subtract the completed hours from the original estimate and then you have Remaining work hours anyway?  No.  The only way that would be true is if all the original estimates were perfect from the start.  That simply isn’t true, and contrary to Agile thinking where you expect change, embrace it actually. 

Statements like “I am 80% done” go away.  Any experienced Project Manager will cringe at the sound of that statement as we all know the PM truism that it’s that last 20% that seems to take longer than that first 80% reportedly done did.  The remaining hours are all you hear about and doesn’t that really tell the story you want to know anyway, mainly: how long until the work is done?

A big question I have heard regarding this switch is: “If we don’t track actual hours, how do we ever know if our estimates are getting better?”  Scrum’s answer: the Velocity metric will take care of that.   It really isn’t about being ultra accurate estimation wise, but instead it’s more about sizing tasks well to allow for achievable Sprint goals when picking tasks to commit to for the upcoming sprint.  Tracking actual completed work hours does not directly help with that goal what so ever.

When the team switches to reporting just Remaining work hours, they also need to switch their mentality to one that has them reevaluating how much work is left on the task at least once a day and updating the task to reflect this info.  It’s important to stress that the tally can go up… it’s not a number to decrement only (an old habit of completed hours reporting that can be hard to break).

Next up in this series, the importance of making it E-A-S-Y for the team to update Remaining work hours at any time.

For both Sprint planning, and Product planning, teams I have been part of have found that being able to see (visually) the backlog is very important. 

If you are using good old fashion index cards/post-its to record your User Stories and tasks, or using a electronic tool to keep track of them (maybe you have a distributed team), all team members need to be able to see the Sprint / Product backlog and should be able to look at each anytime. I would go one step further and say that you also need to be able to visually order items in addition to being able to see them.

Providing the ability to the Product Owner to visually prioritize User Stories is a must in my opinion.

To do this, all User Stories need to be easily viewed, side by side.  Teams I have been part of have used software to track our User Stories and Tasks because we have some team members that are not at the same site as all other team members.  We have tried to have the P.O. prioritize User Stories without a visual tool, and it simply requires too much memorization to do efficiently/effectively.  Once we implemented a tool that allowed the P.O. to see all items on screen, next to each other, as well as drag and drop them to quickly prioritize, the process became much more streamlined and we were able to focus on the tradeoffs between User Stories instead of trying to remember what we set to what priority previously.

The Team needs to be able to visually see the backlog to be able to efficiently prepare for Sprint Planning.

This point is more a call to ensure that all team members have access to look at the backlog at their convenience.  This empowers them to be able to prepare for upcoming planning meetings, as well as get a overall feel for what the P.O. currently feels are important User Stories to tackle in upcoming sprints.  I add the ‘visual’ part to this statement for similar reasons as the P.O.’s need.  Being able to see work items side by side is powerful from a relationship basis.

It’s imperative for the team to be able to see all User Stories while executing Sprint planning.

The team members need to see all User Stories side by side when committing to individual User Stories for the Sprint.  If the team isn’t allowed a full view of choices, they will have difficulties choosing the best stories to commit to given the Sprint goals.  Certainly the highest priority User Stories will be shown first, but those aren’t always the stories the team chooses to implement (for a number of reasons outside the scope of this post). 

Next up in this series, the importance of tracking Remaining Work Hours as opposed to completed hours.

I have heard of many groups that will finish Sprint planning by taking all tasks committed to by the team for the sprint, and assigning them out to each team member.  Teams I have been part of have tried this (especially the first few sprints when new to Agile planning) as it felt like the logic last step of planning.  How else is a team member to know what they would be working on over the next two weeks (or whatever sprint timeframe was chosen)?

The opposite would be to not assign any tasks to individuals during the planning meeting, but instead have the team members select tasks as needed.  Having tried it both ways, I would submit the following:

Best Practice: Don’t allocate all Sprint backlog tasks to individual team members at end of Sprint planning meeting, but rather have team members pick tasks from the Sprint backlog as they come to the point of needing more work during the Sprint.

This seems counterintuitive to most, myself included.  Questions flood to mind like:

• How will we be sure that the ‘right’ people do the ‘right’ tasks? 
• Aren’t there specific tasks that only the ‘experts’ on the team should do? 
• What about situations where someone has specialized knowledge regarding a specific task or set of tasks? 

Not only questions regarding task allocation, but also about individual commitment and getting the work done come out. Not having a single person responsible for each task brings questions like:

• Who do we look to for assurance each task will get completed?
• Doesn’t this open up opportunities for some team members to slack since they aren’t individually responsible for specific tasks?
• What about those team members that need to feel the time pressure of a number of tasks in their work queue to work at optimal speed?  (Many of us work best ‘last minute’)

Based on my experience, and that of others I have spoken with, many of these questions are more pertaining to the cultural shift from traditional Project Manager led teams, to self organizing Agile teams.  Though you may have the fear that some people might slack, or tasks might not get completed during the sprint because they isn’t a single person responsible for each task, the actual opposite happens. By leaving the tasks in the sprint backlog to be picked from as the need for more work arises, the team collectively takes on the responsibility to complete all tasks, instead of just ‘my’ tasks in the other scenario. This has proven to be a much more efficient way to enable the team to ensure they complete the sprint goals and do it in highest priority order.  Also, the daily standup meetings help ensure all team members are ‘pulling their weight’, as there simply is no way to hide if you are updating your team members each day regarding task execution/completion.

Certainly there are tasks that naturally go to certain team members based on their past experience/etc.  But more often than not, the team members know of these situations too, and therefore it works out just fine and in those rare instances that it doesn’t, the team works out the situation as it’s identified and grows from the experience.

To ensure we consciously search for situations that could get us in trouble task allocation wise, teams I have been on have gotten into the habit of asking at the end of Sprint planning: “Are there any tasks in this Sprint that someone feels they MUST specifically work on?”  Very rarely do we identify such tasks.  When we do, we discuss as a team the reasons and decide if the task should go to that person right there.  Otherwise, once the planning meeting is over, each team member selects a task (from their desk) they can ‘Race to Done’ on, and enlist the help of others on the team if they need such help.

For those people that work best ‘last minute’ and feel the need to have a large task list assigned to them, they can simply shift their focus from an individual work queue to the team queue and race to done.

The team can use the Sprint Burn down chart to measure how well they are doing completing the Sprint tasks/goals and adjust accordingly if the slope of the Burn down is showing Sprint goals won’t be completed.

Use of the Burn down chart enables the team as a whole to measure progress and adjust as whole while the Sprint is progressing.  This in conjunction with the Daily Standups keeping communication open and bringing issues out quickly, really does work well and eliminates the need to assign all Sprint tasks at the beginning of the Sprint.

Next up in this series, the importance of Visual tools for Sprint planning.

The best way teams I have been on have found to estimate via a group is Planning Poker.

There are many benefits related to planning as a group (see my previous post on this topic for more).  One thing that was difficult to grasp for me at first, is the idea of group estimating.  Have everyone estimate each task? Even tasks they know they aren’t going to be responsible for?  Yes! The collaborative nature of group estimating helps further dig up hidden features/assumptions as well as provides other benefits. 

Teams I have been part of have tried a number of different ways to group estimate like:

• Group determines (informally) who they think the task is going to be completed by, and defers to that person’s estimates.
• Group members all write down a hour estimate on a piece of paper, let everyone know what they thought, and then negotiate until some consensus is gained.
• Play planning poker for estimation.

All were used to some success level, but the first two generally took much longer to complete the estimation process and also reduced group ownership of all tasks to more individual ownership of certain tasks.  In one instance, the negotiation process for one estimate (not the details about the estimate) took over 30 minutes, and even after that time, we just took the highest estimate to be able to move on.

Looking for a better way, the team discovered Planning Poker.  Here are the high level details:

• Each team member has a set of ‘cards’, each with a single number.  We use a variant of the Fibonacci sequence for card values (1,2,3,5,8,13,20,40,100′).
• A item to be estimated is read to the group.  Any team member who has questions about functionality/etc. are encouraged to ask their question(s). This continues until all questions are answered the best they can be.
• The group facilitator asks each team member to pick a single card.  (Says: ‘Estimate!’)  The card is not shown to other team members at this point and verbal estimates are avoided to reduce the chance of influencing other team members.
• All team members estimate (minus the Scrum Master and Product Owner if they are in the room)
• Once all team members have a estimate card, they are all flipped over so all can see. 
• When there is a significant variance between estimates the person with the highest estimate and also the lowest estimate are asked to briefly explain why they picked the number they did.  This usually exposes differing assumptions by each team member and allows for some quick discussion on which assumptions are valid.
• The team is then asked to re-estimate. 
• This process if repeated until there is group consensus.
• Side note: This Planning Poker in detail page has a detailed outline of the process if it’s totally new to you.

A few tips/takeaways from my experience to help with playing Planning Poker:

• It rarely takes more than two or three rounds of estimating to have total group consensus.  Yes… Really! 
• Invariably when starting group estimating in general, team members will ask if they can ‘abstain’ from estimating some tasks.  We have chosen as a team not to allow this.  All team members need to participate and do the best that they can.  This has helped all team members get better at estimating tasks they usually wouldn’t be asked to participate in as opposed to relying on specific experts and disengaging.
• If there is a small variance in estimates (3/4 of team picked 5 and 1/4 picked 2 for example), the team will discuss amongst themselves what they think is best and usually pick a number instead of doing another round of estimating.
• As a team, we have chosen to stick with estimates that correspond to numbers on cards.  This eliminates the tendency to say: “you have 5, I have 2, lets just average them out to 3.5 and use that estimate”. 
• We eliminated the 1 and 3 cards, as we found the difference between 1,2,3,5 to be too small to be concerned about as a team.  This further encourages ‘bucket picking’ even at the low end of estimating to validate that we are really just sizing activities, not committing to specific time frames. (This decision has been debated a few times since in sprint retrospectives, asking to put the 1 and 3 back, but the team has decided to keep them out for now.)
• Have a way to limit question/discussion length prior to estimation.  Some times team members can get caught in the details and ramble on for quite some time.  I have seen some groups use a 2 minute timer that any team member can start, to limit the current discussion point of which when the timer is done, a round of estimation is required, keeping the process moving. Point here is: group is not trying to precisely estimate the tasks, but instead they are sizing them and that just needs to be ‘within the ballpark’ as the main value is gained in that short amount of time.
• Consider a ‘I need a break’ card.  Some Planning Poker decks have a picture of a pie on one card, meaning ‘I need pie!’ When this card is shown, the group takes a mini-break.  Group estimation can be quite taxing, so breaks are important to keep people fresh and avoid the pitfall of ‘lets just get this done’ estimation.
• Like most things, the first few times your group uses Planning Poker, it will take longer to get consensus, but over a number of sprints, estimation goes much faster as the group gets comfortable with the process in general.

Next up in this series, my take on how to best manage Sprint backlog task allocation to team members.