Skip to main content


What if your todo list is part of a distributed, unsupervised task management app that's running your company without anyone realizing it?
Published onJul 28, 2021

In Wired, Clive Thompson has a new piece about the psychology of todo apps that, if you’re anything like me, will make you feel very seen, as the kids say. Long story short:

This is the black-metal nature of task management: Every single time you write down a task for yourself, you are deciding how to spend a few crucial moments of the most nonrenewable resource you possess: your life. Every to-do list is, ultimately, about death

To reach this conclusion, Thompson interviews psychologists and todo software makers, in the process making clear something I think we’ve all known all along but have been too ashamed to admit: task management software’s purpose isn’t to help us get things done, but, literally, to help us sleep at night.

Thompson posits that todo apps mainly help us avoid the Zeigarnik effect, the mind’s tendency to fixate on unfinished tasks, by giving us a place outside of our brains to stash all of our unfinished business so that we can sleep at night. Thompson even cites a study that concludes people with todo lists fall asleep faster.

But that doesn’t mean we ever finish that business. As Thompson notes, even the makers of todo apps, who track their users’s task completion, admit that in the best case, fewer than 50% of tasks ever get checked off.

So what’s going on here? Thompson gets at the root cause in an almost-embarrassed parenthetical. As I’ve been thinking about a lot, it’s not our fault. It’s, well, read for yourself:

(We could snap the lens open even wider and have a fuller reckoning with capitalism. Focusing on our individual ability to tread water—with apps and lists—can look like a bleak exercise in blaming the victim, when in reality the only solution is not better apps but non-hideous workloads, debt relief, and a saner landscape of civic care. Frankly, if you took “managing grotesquely useless and bloodsucking for-profit health insurance” off people’s to-do lists, it would remove one remarkably stressful item, as my Canadian upbringing compels me to suggest. But I’m writing this particular article from within the belly of the whale, as it were.)

Bingo. My own Cal Newport-inspired “belly of the whale” take continues to be that the knowledge economy’s relentless focus on optimization, productivity, better returns and so on operates not by realizing any actual workplace efficiencies, but by foisting more responsibility on a smaller number of individual workers. For instance, managing a todo list and getting the list done. Factory workers aren’t using todo apps, because their next task literally comes down the line at them. Someone else decides what comes next. They just have to do it.

I’ve long suspected that most knowledge workers effectively operate by foisting responsibility for what comes next onto...something else. Unlike a factory, though, that system is far from coordinated. In fact, I don’t think most people even know it exists.

Per the stats Thompson cites, most people spend a lot of time offloading their todos into task management systems only to complete less than half of them and eventually declare task bankruptcy. That completion rate is bad enough on the individual level, but I suspect it’s much worse when you consider organizations as a whole. Think of how many times you’ve been asked by your boss or a colleague to complete something they weren’t able to. In effect, they’re using you as a human todo app, getting another task out of their brains so they can sleep at night and delay their own task bankruptcy, whether or not they actually expect you to get it done. Which means that much of what you end up doing, whether you realize it or not, isn’t dictated by you, but by a tangle of other priority systems.

In a very real sense, this means most companies operate on a completely unsupervised, distributed todo app. When a task gets assigned to an individual employee, it goes into their individual system, where it has a less than 50% chance of actually getting done. The remaining 50% of tasks are either given up on or passed to another employee, where the same coin flip happens again. Each individual employee’s system’s efficacy, their ability to judge priority, their productivity, and the amount they delegate are all variables that determine what gets done.

Compare that system to how a factory works. It’s completely insane, and a miracle that anything gets done at all. Of course, knowledge work isn’t factory work, and I’m not advocating for strict central planning, which, at least in software development, doesn’t seem to work that well either. But I have to think there’s a better way to do things — or at least, a better way to understand what’s actually happening inside the black box that seems to be running things.

No comments here
Why not start the discussion?