This week we talked about three other thieves of time: unknown dependencies, unplanned work, and conflicting priorities.
Thief of Time: Unknown Dependencies
- Dependencies can be architecture, expertise, or sequential activities.
- If you’re the only skillset expert, you are a bottleneck. This doesn’t necessarily have to be the smartest or most technical person: It could be a role, like QC.
- The book presents an interesting look at the likelihood of delivering on time. If there are five things that all have to be on time for the whole to be on time, there are 31 ways for it to be late and only one way for it to be on time. We thought this was compelling, but too black-and-white. It doesn’t account for other options (e.g., can we deliver something less or in a different time window) and doesn’t factor in probabilities (i.e., something is binary — either late or on-time).
- Relating to the previous concept, it feels like this is biased toward a worst-case scenario, and that failure/lateness is inevitable. The math is accurate, though. The goal is to remove dependencies to increase your odds of delivering on time.
- You know your time is being stolen when
- Coordination needs are high
- People aren’t available when needed
- One change unexpectedly changes something else
- Small teams that communicate can move faster, but you may pay the price of moving fast as an organization. Jameson gave an example from a previous company where two product teams ended up building the same product because neither team knew about the other.
Thief of Time: Unplanned Work
- Unplanned work is firefighting, just like in The Phoenix Project.
- Encourage responding to change over following a plan.
- This chapter reminded Geoff of the VUCA concept…
- Volatility – unexpected, unstable, unknown duration
- Uncertainty — change is possible but not a given
- Complexity — interconnected parts, volume of work
- Ambiguity — causal relationships are unclear
- The more unplanned work you have, the less time you have to do the planned work.
- Bugs are a form of unplanned work.
- It’s unsustainable to do planned work plus unplanned work. How do you handle this? Usually throwing more bodies at the problem is the answer, but then you get into the Mythical Man-Month. You need a good project manager to lead things.
- The team needs agency to pump the brakes when needed, otherwise technical debt creeps in. Geoff’s former tech lead called this “slowing down to speed up.”
- Jameson shared a story where leadership asked people individually to list things that would make their lives easier. This is a neat way to surface where extra friction is happening, typically related to unplanned work.
- Geoff shared an example where the team got a dedicated firefight sprint to address a handful of performance problems without having to focus on other features or other company meetings.
- Reserving capacity of unplanned work can be helpful. Houston has implemented this in a previous project by adding “buffer tasks” to the backlog. This also helped him figure out how much unplanned work was happening each sprint (instead of padding existing tasks higher).
- (Getting a bit ahead, we discussed physical versus digital Kanban boards. It seems that physical boards have more gravitas, as managers can walk by and see them; also, moving physical sticky-notes makes it more tangible.)
Thief of Time: Conflicting Priorities
- Something that can lead to confusion is having multiple systems where information is stored (e.g., wikis, calendar invites, notepads). Having a centralized place solves this problem. (This is one of the fundamental tenets of Getting Things Done.)
- You need information radiators to make the information come to you (push-based).
- Even if you have a system, now the challenge is organizing it and the discipline of maintaining it.
- If everything is important, nothing is.
- We get into trouble by trying to do too much at once. Management needs to line up the priorities. Capacity planning lets you know if you’re busting your limit. Try removing things using “yes, but” — e.g., “Yes I can work on feature X, but I won’t be able to finish Y.”
- Shifting priorities is typically indicative of a leadership problem. People need to know what to focus on.
- We talked about how individuals that are split across multiple teams/projects (i.e., matrixed organizations) feel conflicting priorities because they have multiple bosses, so to speak. One approach is to schedule dedicated time for helping that team/project.
- Conflicting priorities leads to other time-sinks like too much WIP.
- “There is one most important thing — let people know what that is.”
Question for next time
Which of the five time thieves seems the most severe?