Working as a team of internal coaches isn’t easy. We do complex work with a wide range of people, across many areas and levels within the company.
A recurring challenge is planning and committing to work. With so many stakeholders, and so many things we could be doing, how do we draw up an effective plan without spending a lot of time on it?
We’ve embraced weekly planning sessions, and we’re edging closer to a system that really works for us.
Step 1: Are We Winning?
Like many teams, we aim to achieve certain objectives and judge progress with some key results.
The first step of our planning sessions is to review those key results:
- What impact did last week’s work have?
- Are things trending in the right direction?
- Are there signs we want to pivot, or even completely change direction?
Step 2: What Could We Do?
We maintain a backlog of tasks. It’s a roughly month-long plan of what we think we’ll do, which gets reviewed in every planning session.
There’s no formal sizing of these tasks, but we do break them down into chunks that “feel” small enough to do.
Once we understand where we are against our objective, we clean this backlog:
- Review new tasks. Do they contribute to our objective, and do we all understand the task? Is it something we should do? Anything we can’t agree on gets removed
- If anything doesn’t align with our goals anymore, or we think we won’t do it, it gets removed
- Make sure everything is prioritised to help us achieve our objective
Crucially, it’s one single list. No separate backlogs for work from different sources, for rework, for bug fixes. Everything gets prioritised together, to show what we should do first.
Step 3: What Will We Do?
Once we have a prioritised backlog, we commit to how much we can achieve that week.
Until recently this meant a long discussion about the work, what we could or couldn’t do, and committing to tasks one at a time. We weren’t getting the benefits of managing our backlog.
These days, everyone on the team secretly writes down a number. That’s how many backlog items, starting from the top, that person thinks we can complete that week (considering the tasks, holidays, etc.).
We then reveal our estimates at the same time. So far we haven’t varied by more than one or two items, and outliers have always been happy to go with the majority.
That 15–30 minute discussion now takes closer to a minute.
The biggest problem now is getting all those tasks finished. Even when we have a good understanding of the tasks, we often have mid-week discussions about “when can we call this thing ‘done’?”.
We’ll probably experiment with intentional “Definitions of Done”, especially for our more complex tasks, to help with this. Hopefully we can report back on that soon!