Thinking tool: Sometimes you just have to grind

Headshot of Adam Jones

Adam Jones

You can be much more productive with the hours you have by working smarter.

But at some point you hit the brutal fact: you will be more productive if you grind harder, either by:

  • working more intensely during the hours that you work
  • working more hours

This feels uncomfortable to say: it doesn't feel "nice" to work hard, long hours. But I think it's probably the truth.

I used to believe that you'd hit diminishing returns quickly after a large number of hours per week, and that there would be decent evidence for this.

However, I spent time looking at this and ended up changing my mind. There's surprisingly little solid evidence for there being diminishing returns to working hours, and the few robust studies we have suggest working more hours is productive until fairly extreme points e.g. becoming physically unable to continue due to hand cramp1 - although these studies themselves are primarily focused on physical manufacturing, so it's hard to say how they would translate, if at all, to mental labour. It really annoys me that there aren't great studies here, even slightly contrived ones!2

That said, before people recklessly take this to extremes:

  • Being maximally productive at work is not everyone's end goal in life. Although productivity is good: there are a lot of important problems in the world that you can fix! (And if your work isn't fixing them, consider changing that)

  • It seems anecdotally like different people have very different limits on how much they're happy working. That said, these thresholds are probably flexible and can be extended by e.g. thinking about how to make your work more fun.

  • There are some physical limits you will run into in the extremes (e.g. extreme sleep deprivation, repetitive strain injury), although I think most people are far away from these solely from a work perspective.

  • On burnout,3 my theory is that long hours can catalyse bad outcomes from values conflicts (e.g. long hours = kindling, values conflicts = match). And therefore long hours can be quite high risk, but long hours alone rarely cause burnout.

    Work-related values conflicts are probably most often caused by not finding work meaningful. But if you truly enjoy the hours of your work, it becomes a lot easier to work more hours with fairly low risk of burnout. (Making yourself like your job, and making your job more likable, are things you probably have agency over!)

    Finally, some people have values which cannot be fully satisfied by work, e.g. valuing family time. These require spending time in ways that usually trade off against work, and not doing this will cause burnout. But again I see this less as a "too much work" thing and more a "not enough family time/personal time/rest" thing; although of course the two are interrelated and sometimes trade off against each other.

Per MC Hammersmith, Fully Qualified Gangsta (NSFW):

How am I so sick, I put in the work
Why am I the greatest, I put in the work
How do I do it, I put in the work

This is part of my thinking tools series. Also consider:

  • More dakka: more hours is more dakka, applied to your time
  • Defend slack: you can work extremely hard and still keep slack
  • Run the experiment: if uncertain about all this, you still could try working many more hours for a week or two and see how you find it!

Footnotes

  1. The best data comes from British munitions factories in World War 1: the government tracked detailed output records while workers' hours varied enormously, and researchers like H. M. Vernon analysed them for the Health of Munition Workers Committee. ↩

  2. When I bring this up, people often point me at four-day work week studies. I've looked at a lot of these, and almost all suffer from:

    • having major confounders e.g. in both the Microsoft Japan study and the UK companies study, they spent several weeks optimizing business processes immediately before the period, and during the period substantially changed how they would work, to cancel many meetings etc.
    • measuring productivity poorly e.g. in the UK companies study they measured revenue, but not against any kind of control group, so it's hard to know whether businesses did substantially better/worse than they otherwise would have. If it turned out they were 20% less successful, I wouldn't be surprised: work 20% less, get ~20% less results.
    ↩

  3. Burnout is another thing where I'm annoyed that we don't have as much good science on it as we should. ↩