The line I was proudest of, for about 14 months, was that I ran a product with zero meetings. No standups. No syncs. Not one calendar invite. I said it the way people announce they have quit caffeine: a little smug, waiting for the flinch of envy.
Then I put one meeting back. A single recurring call, once a month. And the month after I added it, I closed a decision I had been circling for three weeks. The timing nagged at me enough that I went back through my own notes to work out what the meetings I had so proudly deleted were actually doing. The answer was not flattering.
TL;DR
- Zero meetings as a solo dev was half right: the coordination meetings I deleted really were dead weight, because there was nobody on the other side to coordinate with.
- It was half wrong. Meetings also force decisions, put a witness on your commitments, and drop a hard stop on the workday. Those jobs did not vanish when the invites did. I just stopped paying for them on purpose and started paying interest instead.
- The fix was not reinstating meetings. It was rebuilding those specific jobs deliberately, with one standing call, a timeboxed weekly review, and a habit of deciding in public.
The boast, and where I picked it up
I did not invent the anti-meeting position. I inherited it, fully formed, from people I respect.
Paul Graham's 2009 essay "Maker's Schedule, Manager's Schedule" taught a generation of builders that a maker and a manager run on different clocks, and that a single midday meeting can blow a crater in an afternoon by splitting it into two pieces too small to build anything in. A year later, Jason Fried and David Heinemeier Hansson gave the movement its bumper sticker in Rework, in a section literally titled "Meetings Are Toxic," where they pointed out that a one-hour meeting with five people is not one hour, it is five. I read both in my twenties and treated them as settled law.
So when I went solo, dropping meetings felt less like a choice and more like finally getting to practice what I had preached. And here is the part I still stand behind: for a solo developer, most meetings genuinely are waste. A standup coordinates handoffs between people. A sprint planning session negotiates capacity across a team. A status meeting exists so that eight humans can build one shared picture of where things are. When the team is one person, all of that machinery is spinning against no load. It is the same reason I eventually stopped using an issue tracker built for a team I don't have: the tool was solving a coordination problem I no longer had.
That much was right, and I want to be clear about it before I take the rest apart. Deleting coordination meetings cost me nothing, because there was nothing there to coordinate.
So what were the meetings really doing?
The mistake was assuming coordination was the only thing meetings did. It was the most visible thing, so I concluded it was the whole thing. It wasn't.
When I traced my worst weeks back to a cause, the same few functions kept showing up, and every one of them was a job a meeting used to do off to the side, while everyone thought the point of the meeting was the agenda.
| What a meeting also did | My solo substitute | What it actually cost |
|---|---|---|
| Forced a decision by a hard date | "I'll decide once I've thought it through properly" | Decisions drifted for weeks; a deadline had been doing work I never credited it for |
| Put a witness on my commitments | Kept the plan in my own head | I quietly abandoned plans nobody had seen me make |
| Surfaced disagreement in real time | Argued both sides with myself on paper | I shipped confident mistakes with no one to catch them |
| Dropped a hard stop on the day | Worked until I drifted off it | The task sprawled to fill every hour I gave it |
None of these are coordination. All of them are load-bearing. And unlike the coordination overhead, which I could delete for free, these four were structural. Removing the meeting removed the structure and left the need standing there with nothing to lean on.
The three weeks I lost to a decision nobody forced
Here is the one that finally got my attention, because it was expensive and entirely my own fault.
I had a schema change to make. Small, reversible, the sort of thing that in a previous job would have come up in a Tuesday sync, gotten fifteen minutes of mild argument, and been decided by Wednesday because the meeting simply ended and a choice had to exist by then. On my own, with no meeting to end, the decision had no deadline attached to it. So I did the responsible-sounding thing and kept thinking about it.
I thought about it for three weeks. I wrote two documents weighing the options. I felt productive the whole time, because writing documents feels like work. But the schema did not change for 21 days, and the feature sitting on top of it did not ship, and the actual amount of new information I gathered in week three versus week one was close to zero. I had not been deliberating. I had been marinating.
A meeting would have killed that decision in a day, not because meetings are smart but because they are over at a specific time, and a decision that has to exist by the time a meeting ends tends to exist. I had thrown away the deadline along with the small talk. This is the thing the anti-meeting essays are quietest about: for someone like me, the meeting was never the interruption. The meeting was the forcing function, and the calendar was doing my executive function for me.
What I changed without reopening my calendar
I want to be honest that this is where my zero-meetings boast became literally false, and I think that is fine. The goal was never zero meetings. The goal was zero waste. I had confused the two.
I rebuilt those functions on purpose, and only those. The forcing function and the hard stop came back together, as clocks. A standing weekly review, 30 minutes with a timer running, where any decision I have been "thinking about" for more than a week has to resolve or get explicitly killed, plus a fixed end-of-day I hold to so the work stops sprawling into the evening. The witness came back as deciding in public. When I commit to something now, I write it somewhere a reader could see it, because a plan with a witness behaves differently from a plan I am keeping in my head, even when the witness is a handful of strangers who will never follow up.
And the disagreement came back as the one real meeting, the monthly call, with another solo developer who is under no obligation to be nice to me. I had already written about losing the one person who tells you you're wrong when you ship alone. The meeting is where that person used to live. Rebuilding a thin version of them was worth exactly one calendar invite a month.
Notice what I did not add back: a standup, a status meeting, anything recurring whose job is coordination. Those stayed dead, because those really were waste. I paid to restore the functions I had lost and refused to pay for the ones I hadn't.
Isn't this just meetings with extra steps?
This is the fair objection, and I have argued it against myself more than once, so let me put it at full strength.
You could say I have simply reinvented meetings, badly, and that my "timeboxed review" is a standup with a party of one, my "deciding in public" is a status update, and my monthly call is just a meeting I have decided to feel clever about. There is a version of this where I spent 14 months taking the machine apart only to rebuild most of it and claim I invented something.
I think that objection is about 60% right, and the honest 40% is this. The reconstructed versions are strictly weaker than the real thing at exactly one job, disagreement, and that weakness is not fixable. When I argue both sides of a decision on paper, I am still the one holding the pen, and I cannot fully surprise myself. The monthly call helps, but it is one hour against a whole month of me agreeing with me. I do not actually know yet whether that call is doing real work or whether it is a comfort blanket I have talked myself into needing. Ask me in a year. Forcing, witnessing, and stopping all rebuild cleanly, because they are mechanical and do not need a second brain. Disagreement is the one that genuinely wants another human, and it is the one I am still faking.
Where I think I'm wrong next
If the pattern holds, the belief I am currently too proud of is that async always beats sync. I have spent this whole post defending a slower, written, meeting-light way of working, and I still believe it for most things. But I have started to notice a specific class of problem, usually a tangled decision with three or four interacting unknowns, where 20 minutes of live back-and-forth with one person gets me further than a week of careful async ever does. I suspect I am still underpricing synchronous bandwidth for exactly the decisions that are too knotty to write down cleanly, and that in a year I will be back here explaining which half of this post was wrong.
That is the actual lesson, if there is one. Not "meetings good" or "meetings bad." Just that when you delete a ritual, you should figure out every job it was silently doing first, because some of those jobs do not have the decency to announce themselves on the way out.
A couple of questions I keep getting
Doesn't a weekly self-review just become another thing you skip? It did, until I put a hard external date on it by tying it to the same day I do my monthly numbers, which I will not skip because money is involved. Attaching the soft ritual to a hard one was the only version that survived.
Why not just get better discipline and decide faster without any of this? Because "get better discipline" is the plan that has failed every single time I have tried it, and building a structure that makes the right thing happen on a schedule is the plan that works. I would rather engineer the forcing function than keep losing to my own good intentions.
What if you had employees? Then most of what I deleted would come straight back, because it would be doing real coordination work again, and this entire post would be irrelevant. That is the point. The right number of meetings is a function of how many people have to stay in sync, and for a team of one that number is small, but it is not zero.
Here is the question I would actually like answered: when you removed a meeting, a standup, a weekly sync, a recurring call, what job did it turn out to be doing that you only noticed once it was gone? I am collecting these, because I suspect forcing, witnessing, and stopping are not the only three.
I'm a solo developer building an iOS app called Simple Memo. It turns a thought into a sent email before I can talk myself out of it. I write here every few days about the parts of working alone that nobody warns you about.
Top comments (0)