A small case against systems

Most of what is sold as a system is really a costume for anxiety. I have worn most of them. Here is why I stopped, and what replaced them.

There is a whole industry now dedicated to helping people who have already read one book on productivity read a second one. It is a good industry. The books are, mostly, well-meaning. Some of them are even useful. But I want to describe, in this small essay, a suspicion I have come to hold about the whole category — and about the word system in particular.

Here is the suspicion, plainly: most of the systems I have adopted, and later abandoned, were not really systems. They were costumes. Their purpose was not, in the end, to help me get more done. Their purpose was to help me feel like a person who was doing the kind of thing a person who gets a lot done would do.

What a system actually is

A system, properly speaking, is a set of rules that tells you what to do next in a situation you have not yet encountered. A traffic light is a system. An accounting workflow is a system. Double-entry bookkeeping is a beautiful, four-hundred-year-old system.

What most people, myself included, call a "productivity system" is usually something quite different. It is a shared vocabulary and an aesthetic — a set of nouns you can use with other people who use the same nouns, and a visual style you can apply to your notes. This is not nothing. There is comfort in a shared vocabulary. But it is not, mostly, a machine that produces work.

A system tells you what to do next. A costume tells you how to look while you decide.

The costume test

There is a small test I now apply to any new "system" I am tempted to adopt. I ask myself two questions.

• If I applied this today, would it change what work I finished this week?
• Or would it just change what my desk looked like on the way to finishing it?

Most of the frameworks I have flirted with fail this test in an interesting way. They pass the second question — my desk would definitely look more considered — and they equivocate on the first. And when I have been honest with myself about my own past, the honest answer is that the times I have produced my best work have not been the times when my system was most elaborate. They have been the times when my desk was clear, the deadline was close, and the question what do I need to do this afternoon could be held in a single unaided breath.

What I use now, which is almost nothing

I use a paper notebook, a shared calendar, and a folder of one-page briefs, one per active project. That is the whole thing. There are no tags. There is no board. There is no daily review template. There is a page for today, a page for tomorrow, and a brief for each thing I have said yes to. The rest is judgement, and there is no software for judgement.

I want to add one caveat, because I am not writing this to be provocative. If you work on a team of forty people and you all use the same board, please, for the love of everyone, use the board. Systems that coordinate humans are real systems. They exist because more than one person is trying to remember the same thing. What I am writing against is the personal, single-user system that is really a very expensive way of being alone with your anxiety in a nice interface.

What replaced the systems, honestly

Two things replaced them for me, and they are boring.

The first is deciding, in advance, what I am going to do. Not in the abstract, and not in the language of quarterly objectives. In the specific, small, tomorrow-morning sense of this is the sentence I will write first.

The second is finishing things. Not all things. Not perfectly. But enough things, often enough, that I know from direct experience what a finished thing feels like, and can therefore recognise, in the middle of any given afternoon, whether I am moving towards one or away from it.

Neither of these is a system. They are just the two things a system was supposed to help me do, unbundled from the costume.