2023 W9

We tried out a fairly different way of working this week. Instead of agreeing on a set of things we wanted to build and then going off and doing that, we each picked a topic we wanted to explore and then for 83 minutes each day wrote/drew/thought about it. And then we stopped and that was (ostensibly) all the work we were going to do that day.

Here are my motivations for these parameters:

• Why 83 minutes: I wanted an amount of time that felt right, more than an hour, less than two, and thought round numbers are kind arbitrary.

• Why only one session: Two reasons. First, because I think we needed a rest! We've been doing week long sprints for 5 weeks now. Second because we each picked large complicated topics and there's a common failure case where we get caught up in specifics. A single session let's us focus on it, but then gives our brains time to process in the background.

• Why not a proper full break: I think it's important for us to maintain contact with the work to not feel anxious about it. This may not carry for others, but the theory was it feels more restful to have some contact with the work and then leave it be, than worry about not working all week.

What I was thinking about

I was trying to strongly describe what the app is and what it isn't so that we could apply that to discussions of what we are going to build during a given week. This year so far focus has been a big theme for me. I ended up writing a couple drafts that roughly went like this in length: ▃▇▅▃ . I tried a bunch of different things but one that felt particulary productive was "Pick one", where I try to pick just one thing, one competitor, one differentiator, one use-case.

I think we decided that what we worked on is going to end up published as little memos for us to refer back to, so I'll leave the details for that.

What shall change

I think something like this process is worth it to cap off our week cycles. I am not sure that the "thinking about the single thing again and again" is the right framing, but I think doing a small amount of focused work per day is.

One thing that struck me on Thursday was the difficulty of feedback in this process in particular. If you've been thinking about the same thing intensely for a couple of days, you are going to have so much context cached in your brain that isn't in the artifact that you made. And then if someone else tries to give you feedback based on what they can reverse engineer just from the artifact, I think it's pretty tough on both sides. I think adjusting the process to have each of us "present" our work and explicitly ask any questions we have, could work better. And perhaps just not asking anything, we can happily give each other a couple days space.

Looking back on Cycle 2

Since this is coming in at the end of a 3-week cycle, why don't I sneak in some larger thoughts on that as well, before we get to how the week ended.

Overall I think the increase in velocity is pretty clear. We are building more actually usuable new things than ever before. I think this process of working is well matched to the kinds of problems we need to solve and also to the foundation we have built for ourselves.

It's also, like Brendan mentioned in his note last week, a surprising journey from clarity to confusion again and again. While in the macro we are making great progress, it's a bit of a lottery whether a single day will have a palpable sense of focus. I think that this is, perhaps, a sign of progress, in that at least we have a reference point for what good days feel like, and we touch that reference regularly. On the other hand it definitely points to further advancements neccessary.

One of these is I think shifting from a single one of us "leading" a week, to each of us having different roles, which can shift but do not have to. For example, I've mentioned that Celine is noticeably better than all of us at scoping out a week. She consistently get's a set of work that stretches us just the right amount and feels great to get done. Likewise Brendan is better than Celine and I at structuring something as single activity, whether that is a Space or a single work session. He's also constantly pushing us to run more spaces ourselves, which, though I always have a hard time actually doing it, pays off every single time.

So I think a system where we each fall into the roles we are good at makes a lot of sense. That combined with a different type of cycle to break things up at the end I think can work really well. Gives us a very concrete sense of rhythm, and something to look forward to.

Where we are at

This week our working space was the only active one, and while simple, I think it served it's purpose pretty well. We had a room for exorcising the ideas and frustrations that came to us during our work sessions, and while it didn't work that great at ensuring I didn't think about it after, it was useful to have somewhere to put stuff. And, at the end of the week, there's not that much in there that's radically new. There's a bunch of ideas we've talked about before, but by and large the space did what it was supposed to. This thing, this idea of short term structured spaces, really seems to be working.

On friday, perhaps feeling this, or perhaps buoyed by our week of rest, we agreed that we should launch the app after our next cycle of work. This is scary, but I think it's the right call. It's fascinating how these cycles help us make this decision a lot easier than before, and also how you're not ready until all of a sudden you are. I'm sure we're shipping too late, but it feels good to feel ready to ship.