These are my services: Dailies
This is the first part of a series where I attempt to explain how I see my role as a Scrum Master. Today, we’re going to look into my brain during a daily.
The only mission of each daily I have is to come to an agreement on what is the most important thing to do that day. That’s it. Everything else is discussed outside of the daily.
A common daily looks like the team gathering around in a semi-circle around the task board, with everyone taking turns telling about yesterday, today and tomorrow. I don’t do this, and encourage nobody to do it either.
The boards I work with are structured primarily to help communicate priority and urgency to anyone looking at it at any time.
The board is split into columns with the following logic: The more done a task is, the more towards the right in the board the card is. Only after a task is done, is it valuable for its user. Therefore, the more “almost done” a task is, the more attention should that task receive, to guarantee it getting really done. It’s better to have one thing done, than ten things almost done. The columns are internally ordered according to priority, from top-to-bottom.
My algorithm for completing a daily is roughly as follows:
- Open the daily at the exact time (according to my computer/phone) the daily is scheduled to start and don’t wait for people coming late.
- Start from the rightmost column in the board (I usually skip the columns that contain really done tasks)
- Pick the topmost task, and ask that someone working on the ticket explains how’s it going.
- Repeat down that column, until the bottommost task is status-checked
- Choose the column to the left of the current one, and repeat the two previous steps until we face a column that contains things that aren’t under development.
What I listen for
If nobody talks about a task on the board, I stop the conversation and figure out whether it’s blocked, or if nobody is actually working on it. Blocked tasks happen now and again, and they can be fixed by asking the right questions (usually “what needs to be done to get this moving again?” does the trick). Tasks that get dropped before they get finished is a sign of either A) someone multitasking (bad) or B) work starting without a clear understanding of its importance (worse).
On a general level, I’m interested in whether team is working in a structured manner or not. Indicators against structure are e.g.:
- Everyone working alone on tasks, and tasks aren’t related to each other in any way.
- People having a tendency of having two or more tasks in progress at a time.
- Tasks get no comments or questions from other team members.
I don’t address these concerns during the daily, since the adjustments required will take a rather long time. Having that discussion at this point does not help me in the mission of the daily: understanding what’s important today. Instead, I deal with these things in hallway discussions and retrospectives.
I’m also looking for signs for people struggling alone with problems, thus blocking their progress. While resolving those problems is not the aim for the daily, it is important today to acknowledge those problems. Here I ask people to state their problem out loud, and ask people to help out right after the daily (unless the problem is solved with a sentence or two). This prevents people unrelated to the solving of the problem wasting time (and subsequently getting bored).
Most importantly, if people don’t feel engaged during the daily, my daily isn’t working for those people at that particular time. I would then, right after the daily, ask why it isn’t working, and what would be better. The best course would be to figure out a fix together with the team.
I don’t timebox my dailies to 15 minutes. I do, however, take the time as a very serious guideline, and so I watch the clock like a hawk during the daily. I expect my dailies to be anything from 8 minutes up to 20 minutes, with the average close to the 10 minute mark. After 20 minutes, I still aim to go through all the work in progress, but hurry the discussion up. This is okay, assuming our board is sorted by the left/right and top/bottom value ordering. We already used the most time on the most valuable time, so it’s not a big sacrifice to skim over the less critical stuff.
If a daily goes on for longer than what I expect, I try to figure out the reason. If it’s a one-off thing, I don’t sweat too much about it (but I do make a note of it, and come back to it on the next daily to check whether it still is an issue). But, if it seems like the daily is growing longer, that needs to be dealt with. The actions I choose depend on the particular circumstances.
Since dailies are unique in the sense that it’s the one place where everyone is present and listening, I usually give leeway to some things. Things that would suffer from an abrupt ending, or things that get solved with a sentence or two. But if a topic has gone on for 5-or-so minutes, I interrupt the discussion. I acknowledge its importance and urgency, ask the people needed to resolve it to continue talking immediately after the daily.
Have dailies early
Dailies serve their purpose best if they are early in the day, for several reasons:
- It has the least chance to interrupt someone’s flow, if people don’t have the chance to start working. Many choose to come to work exactly as the daily starts. That sounds like a smart move to me.
- There’s less chance that people would pick up tasks that don’t require the most attention. If there’s a problem to be resolved with another task, people without a tasks are priceless in these opportunities.
- Sometimes something comes up in the daily that affects the priority order. Dailies, being a common gathering, is a perfect opportunity to let these kinds of news be heard. Getting these news early in the morning lets us work on this for a longer uninterrupted period at a time.
Since everyone’s present and listening, I also have a tendency to go through the day’s calendar, reminding of meetings and other events. Or, if there’s some general quick announcements to make, I’ll take the time for that. But only after the daily, and only if the daily hasn’t dragged on for a long period of time. I’ve also noticed an interesting trend where people use this opportunity for announcing their incoming absences, even without being explicitly instructed to.
Feeling of presence
The format I’m used to works best when everyone’s in the same room, looking at the same screen, and seeing everyone’s faces. If you have remote members, using webcams and everyone looking at the same screen (screen share or no) is an alright second-place compromise. Anything less, and there’s so much lost of the communication, that I question the value and/or role of that daily format.
Open doors policy
I don’t consider dailies a secret meeting. I welcome everyone to come and listen. Anyone is welcome to add upon what is already being said during the daily, but I’m careful of off-topic discussion and prolonged back-and-forth dialogue. It’s a delicate balancing act between keeping things concise, and welcoming people to come and interact with us.
Find the benefits
Most importantly, not every team needs to have a daily, if the same need is handled otherwise. Dailies work for my teams, since I have only worked with people working in offices, together, at the same time. I can see that completely distributed teams might have struggles finding a daily concept that would work for them. Decide yourself whether you have or don’t have dailies. But only once you realize that the daily is the perfect solution to the issue you are facing. After a few dailies, verify your assumptions.