Hide table of contents

This is a bit of a sensitive question, but I am curious to know how much other EAs actually work. I think this can easily fall into setting unrealistic expectations in the community. However, I have the suspicion that many people in the community have overly high impressions of how many sustainably productive hours other people make. Also, I'm currently working quite flexible hours and I'm wondering how to count a collection of "actually work" hours as a full work day (I'm currently leaning towards 4-5 productive hours/day). The question I'm specifically interested in is as follows:

If you only track the hours you spend actually doing work*, how many hours do you make on an average work day that is sustainable?

* So this excludes: small breaks, long breaks, distractions, etc.

Additional questions I'm interested in:

2) How many of these hours are deep work?

3) How much total break time do you need? Or in other words: what is your average 'efficiency'? (note: higher than ~75% efficiency seems unattainable sustainably for most people?)

More info is also in this excellent blog post by Lynette Bye.




New Answer
New Comment

13 Answers sorted by

I track my time using org mode. My logs say that I work an average of five hours/day. Since I work every day, that's about 35 hours/week. By "work", I mean time doing actual work, which excludes short breaks to check my email, get a glass of water, or go to the bathroom. I do these ~5 hours of daily work in the course of a 7–8 hour workday.

People use different definitions of "work", and different methods for tracking how much they work. In general, I find that the more hours someone claims to work per day, the more likely it is that they are using a very loose definition of "work", or a very unreliable time-tracking method.

In part for these reasons, I wouldn't pay much attention to how you compare to others, and would instead focus on adopting a consistent definition and method over time. This can help you draw relevant inferences, allow you to make informed decisions, and eventually likely make you more productive, as well as more satisfied with your productivity.

EDIT (2022-01-24): Related post by Erich Grunewald on how he uses org-mode to track his work hours.

I occasionally track my work time for a few weeks at a time; by coincidence I happen to be tracking it at the moment. I used to use Toggl; currently I just track my time in my notebook by noting the time whenever I start and stop working (where by "working" I mean "actively focusing on work stuff"). I am more careful about time tracking my work on my day job (working on longtermist technical research, as an individual contributor and manager) than working on the EAIF and other movement building stuff.

The first four days this week, I did 8h33m, 8h15m, 7h32m, and 7h48m of work on my day job. I think I did about four hours of work on movement building. So that's an average of about 9 hours a day. Probably four of those hours are deep work on average.

My typical schedule is to do movement building stuff first thing in the morning, eg perhaps 7:30am to 8:30am, and then to do my day job between about 8:30am and 7pm, with a 30m break at 10am to hang out with my girlfriend after she's woken up, and a maybe 40m break for lunch at about 12:10. I occasionally do some calls in the evenings, or respond to people's messages about work things. (I usually go to bed between 10:10 and 11pm.)

So my efficiency is probably about two thirds, if you include my morning break and lunch break in the denominator, and 75% if you don't.

I normally work for a couple hours on the weekend, mostly doing calls, and I also usually do some kind of unstructured and unfocused work like walking around and thinking about lots of stuff which sometimes includes my work. So I guess my total work time per week is probably like 47 hours or something.

My efficiency is highest when I wake up unusually early and work uninterrupted for a long time. It's also much higher when I'm doing tasks that it's easy to do for a long time. The most obvious example of this is meetings--they require less concentration than e.g. programming, and so if my day includes a lot of meetings, my efficiency looks higher.

Of course, work efficiency is a dangerous thing to optimize--I actually want to optimize the value of my work output, which is related but importantly different. In particular, sometimes I fall into a trap where I spot some task which I can easily spend lots of time on, but which isn't actually the most valuable. I try hard in this kind of situation to catch myself and ask "what's actually the most important thing to do right now".

My efficiency and total work time has usually been somewhat lower in the past. When I worked at MIRI, I would typically get something like 37 hours of work done in a typical week (roughly 2/3 technical work and 1/3 recruiting work). I also had some bad fatigue problems at various points over the last few years; I think I worked more like 20 hours per week for like a third of 2019, which was very sad and unpleasant. I was kind of depressed for a while last year, which I think took my work down to maybe 30 hours per week. I work more at my current job for a few reasons: it feels more tractable than my MIRI work, I feel more responsibility because I have a more senior role, and I am working more as a manager and so I spend more time doing types of work that I find less tiring.

I think that my current work schedule is basically sustainable for me as long as I feel reasonably happy and satisfied with my life, which is pretty hard for me to ensure.

I can imagine taking other jobs that seemed equally impactful where I'd end up working many fewer hours. And there are a few jobs where I'd end up working more hours (eg jobs where I was constantly talking to people and rarely trying to think hard about stuff).

Incidentally, I think that tracking work time is a kind of dangerous thing to do, because it makes it really tempting to make bad decisions that will cause you to work more. This is a lot of why I don't normally track it.


EDIT: however, it seems so helpful to track it some of the time that I overall strongly doing it for at least a week a year.

I track my time using hourstack.com and try to be quite strict with only tracking 'sit down work time'. I normally can do around 3.5-4h of work a day. I normally start at 10am and finish around 5pm.

This matches my experience at college, where I found I could normally do around 4 hours of studying before feeling tired out.

It's easier for me to 'clock more hours' when I have more meetings. But I try to avoid meetings.

I find that I can get most of my things done within this time and would consider myself a quite productive person.

Can you make this an anonymous poll? (My employers read this). The answers will be more useful/useable.

  1. When I've toggltracked etc I usually get 40-60 hours per week including weekends, but it's very hard to break it down as there are so many distractions.

  2. Probably only 8-10 hours per week of 'real deep work excluding coding'

  3. Maybe 1/3 break time, really hard to know.

I've tracked my time for a few years down to the minute, and usually tracked about 6 hours of "real work" in a day, working about 6 days a week, with also usually an additional 2-3 hours of work on the Sunday. This gets me pretty close to 40 hours. 

However, to get to 6-7 hours of actual work in a day, I tend to spend something like 70 hours in the office per week. If you count time spent in meetings or work conversations as "real work" time, I also tend to get closer to 60 hours a week, though meetings are definitely not all fully focused.

I time track ~30hrs a week.

I get ~6hrs a week of "deep work", mainly because I am in management and a lot of management activities are not classically considered deep work.

My efficiency is ~70%.

I'm grateful you've asked this question, as I've been really curious about this myself and have considered asking the same question. I know I should be careful when comparing myself with others, but I only have experience from one organisation where I have also been in charge of deciding number of hours we should work and how to track it. It feels nice to know that what I've been doing isn't totally off.

Here's what we do at our ~3 year old organisation with 3-4 employees: We use Toggl for tracking hours we actually work and a separate spreadsheet to track number of hours we're at work. We estimate that 6 hours on Toggl is about 7.5 hours at work. The extra 1.5 hours are breaks, usually consisting of many 5 min breaks and one 30 min break (we don't count lunch).

I usually track 6-7 hours of actual work in a regular day ( I work 5 days a week). Some weeks I track 4-5 hours most days, and then 8-10 hours some days. I try to regulate my hours so that each week is more or less the same. In weeks where I work a lot over 6 hours per day, I schedule a day off in the coming week.

I don't have that much time for deep work. Maybe 1-4 hours per week. In my current role, not that much of my work requires deep work, though I should probably be doing 3-5 hours of this per week.

I've tracked my time for a year working remotely doing research and it comes out to between 25 and 35 hours a week. 

I'd guess a little more than half is deep work where I am fully engaged and undistracted. Most of the time this means taking no breaks for a several hour stretch every day. It's not uncommon for at least half of the deep work to be misguided or not best spent on reflection.   

I'm not sure what to imagine when I hear an amount of weekly hours when working remotely. Working 40 hours a week at an office or on a job site can be relaxing compared to the weeks where I track 30 hours or less since it's common to spread five hours of work across a "normal" eight hour work day span of time. 

I next describe what different quantities of hours worked looks and feels like. Basically, my guess is that 1 hour of remote work for me = 1.5 hours of "office work" so 20 (40) hours worked = 30 (60) hours spent "at the office". 

In a 20-25ish hour work week (if not caused by low mood) I typically am balanced and happy, feeling like I have most of my afternoons and evenings free to exercising, see friends and create things. This would be ideal to maintain, and having these weeks keeps me from burning out. (Aside: normally the intensity of work weeks cycles between high and low intensity work weeks). 

25-30 hours. In these weeks I maintain the habits I find essential to keep going, but it feels like just barely. On half the days I finish work, and then immediately go for a run before it is dark, return home to frantically cook dinner then squeeze another hour working before winding down which normally does not include a discrete leisure pursuit beyond listening to a podcast while tidying up the office-house.  On two maybe three if I'm lucky work days I do something that's fun but not exercise for at least an hour. 

30-35. I have one or two periods of time during the work week spent doing something deliberately not work related. My relationships feel a bit strained, if there's a quiet time it's spent in transit or doing chores. I imagine this as hard to maintain, and in the deadline weeks (or god forbid months) where this persists I feel myself wearing thin. 

35-40. If I'm working this much something has gone wrong. There is nothing but work. It feels as if I spend the whole day, every work day engaged in work or thinking about it. I may not leave the house for a couple days. This normally means a few chunks of the weekend slipping back to do something "light" and "easy". Every non-dinner meal (which are often few and hastily prepared) is consumed at my desk which I'm at minutes after waking. There is sometimes a break for dinner, but if I can I'll eat that at my desk too. During these (rare) weeks things start to fall apart. 

These weeks are frequently followed by a hangover week where I crash, and work 20ish hours.  

  1. Looking at my Toggl, I've logged ~35 hours/week over the past few years since becoming self-employed.
  2. However, I log hours for meetings and work travel (and personal stuff like volunteering and online courses), which tend to require less concentration. When I subtract those out it's more like ~25 hours/week of "deep" work.
  3. I only log the hours I work, so it's hard to tell how much of my remaining time counts as a break. But I doubt I'd be more productive at a 40 hours/week job, so call it (~25/40 =) 62.5%.


I'm part-time/hourly, so have been tracking my time for years, "clocking out" for short and long breaks. I've averaged about 6.2 hours/day since the beginning of 2020 (to pick a random period of time). That excludes paid leave (holidays, PTO, sick).

I tracked what I spent my time on for the whole of last year (you can read about this more here), as I finished university and started my PhD. I've always thought that I was working a lot since my course was quite intense, but I found out that on average, I also work around 5-6h per day. To quote the post: 

Finding out how little focused work I do, even on the days that feel productive, was one of the biggest surprises of analysing my data. It turns out that a typical working day (9-17) would usually only give around 5-6 hours of actual work, while the rest of the time would dissolve on low-focus work, chatting to people, coffee breaks and other little distractions. While there were days when I would reach 8 hours of focused work (for focused uni work I had 32 such days this year), they required either spending the whole day (9-22) working or a lot of effort to stay focused. Both were quite tiring and unsustainable in the long run.

For this year, my average number of working hours per week is 33 (this doesn't exclude free days/holidays though).  There is quite a lot of variance, with my last 5 weeks of working hours being: 58, 31, 29, 50, 39. It seems like there are weeks of high intensity (usually caused by a surge of motivation, coding-heavy tasks or an upcoming presentation), followed by weeks of lower motivation & ability to focus. It's interesting that it still averages out to ca 6h/(working) day.

  1. Roughly 40 hours of actual work time per week — 35 at CEA, then a few hours in a few different places (the foundation I assist, editing for friends/relatives, the bits of my streaming that count as "work")
  2. Because much of my work involves writing or copyediting lengthy documents, I have a high percentage of "deep work" (I'd guess at least half my hours), though it's mostly not the fancy kind of deep work where you come up with ways to save time and improve systems. It helps that I don't have many meetings..
  3. I work in spurts throughout the day, rarely more than a few hours at a time (lots of unpredictable obligations right now). But if I were to have eight uninterrupted hours in which to work, I think I'd end up at 5-6 hours of actual work time, so ~60-75% efficiency.

1 Related Questions

2Answer by david_reinstein
Or maybe these are the mediators and not the prime movers. When I am in certain mental states most of the problems that I mention above can become much more severe or they can disappear entirely.
9Answer by Guillaume_Chauvat
I think most of 1 and 2. But also, on a larger scale, something related to the Peter principle can happen: I spend way more time doing things that I'm not motivated for, because those are the ones I get stuck on. The fun parts are over more quickly.
Sorted by Click to highlight new comments since:

[At best a semi-answer to your question. Sorry.]

If I reflect on my last 5 years or so, one thing I find really striking is how much both the quality and quantity of my work has depended on external circumstances.

In particular,  I have experienced all of the following:

  • Working a lot, having a lot of "bad stress", and feeling quite angry and miserable.
  • Working very little, being depressed because I can't seem to find resonance with the world, and developing an exaggerated sense that there is nothing useful for me to do, I can't contribute value, and can't do anything that's good for anyone.
  • [The recent months, and also some earlier periods:] Working a lot, having a lot of "good stress", and feeling great.

During the times when I was procrastinating a lot, generally had low motivation for "work", and was depressed, it was often quite bad for me to confront the fact that other people were (or at least seemed) productive and happy (whether or not they worked long hours).

But I think one of the worst lessons I could have learned is that other people are just intrinsically more hard-working or capable, and that I should leave the job of improving the world with these 'great people'. I also think it would have been bad to approach this as a 'productivity problem' to be solved with 'productivity hacks' or other local fixes.

(This is not to deny that interpersonal differences in productivity or capability exist as well. I'm kind of on the record about this ...)

If I could give my past self some advice, it would be this: Don't compare yourself to others. Treat your depression. Lean into what you're good at. And if your environment isn't good for you, change it. I wasn't great at actually doing these things, but I'm very glad about the extent to which I was.

This is an interesting question, though I am somewhat concerned that the responses will be biased towards high numbers because people who work relatively fewer hours may be less likely to respond. I would give much more weight to an anonymous survey.

On a different note, I have personally found it useful to track my working hours using Toggl Track (https://toggl.com/track/). This has given me a much more accurate sense of how many hours I usually work per week and how long I should expect projects to take.

FWIW, I also think it's plausible responses will be biased towards low numbers because people want to be avoid looking like they're bragging, don't want to contribute to people's stress, etc. 

(But to be clear, I'm not saying I expect those different sources of bias to cancel out - it seems hard to say what the net bias would be - and so also endorse the idea of giving more weight to an anonymous survey.)

Others are answering in units of hours, but in terms of toggling between accomplishing real work and getting caught in distractions, I feel like the more relevant unit for me would be in minutes. I predict that an attempt to log my time “actually working” would be doomed because I flit in and out of “actually working” on a much finer timescale than it is practical to track. Perhaps this is a sign I would benefit from an attention-improving drug.

Hah! Random somewhat fun personal anecdote: I think tracking actually helped me a bit with that. When I first started tracking I was pretty neurotic about doing it super exactly. Having to change my toggl so frequently + seeing the '2 minutes of supposed work X' at the end of the day when looking at my toggl was so embarrassing that I improved a bit over time. Now I'm either better at swtiching less often and less neurotic about tracking or only the latter. It also makes me feel worse to follow some distraction if I know my time is currently being tracked as something else.

Curated and popular this week
Relevant opportunities