This year, I've been doing three month personal reviews (Dec '13 - Feb '14, March-May, May-June) to track my progress against my EA goals.  In this model, I review what I have done over the past three months, where it has gotten me, and then reflect upon it.  In doing this review, I hope to be not all that different from the three month review of any other EA org, except I'm an EA org of precisely one person.

Hopefully, posting such a review here is useful as a case-study of a particular effective altruist approach, and you might find something useful in it, or have some useful commentary to offer me.  I'd also love to see some of these from other people.  (Just sayin'.)

General Philosophy

The overall idea is that I self-identify as an effective altruist and want to make sure that a large bulk of my time is spent on some sort of personally optimal path to EA impact goodness.

More broadly, I've recently taken to heart the idea that since I am pretty young and don't yet have a good idea of what my talents can get me and what my options are, I should try to do a wide variety of things that seem like they could be useful down the road, and then assess them later.  One could think of this as the "explore" mode of an "explore-exploit" model.


How Did I Spend My Time?

On 27 Jul 2014, I switched to Toggl from Google Calendar for my time tracking of choice (more notes on productivity forthcoming).  This means I now have much finer-grained time data, potentially down to the minute, though there are some inaccuracies.  Here is how I spent my time over August and September.

ProjectTotalPer Week[1]% Week
Sleep 482.2 56.3 33.5
Day Job 250.86 29.3 17.4
Social 238.3 27.8 16.5
Break 113.94 13.3 7.9
Prep[2] 107.6 12.6 7.5
Programming 76.6 8.9 5.3
EA Direct 65.9 7.7 4.6
Other 39.4 4.6 2.7
Planning 32.7 3.8 2.3
Moving 30.8 3.6 2.1
Email 26.8 3.1 1.9
Entrepreneurship 19.3 2.2 1.3
Food[3] 12.2 1.4 0.8
Cleaning 6 0.7 0.4
Errands 4.1 0.5 0.3
Exercise 3.2 0.4 0.2

[1]: On average – obviously not all of these activities took place every week.
[2]: Includes getting ready in the morning and all commuting to and from places (not just work).
[3]: Includes making, but not eating food.


What Did I Accomplish?

My Day Job

Besides sleeping, I spent the next most largest amount of time on my day job.

This job is full-time, and supposed to be 40hrs a week or more.  The reason why this appears as under 30hrs a week over the two months is because (a) I took one week off for vacation and (b) working 8 hours of day is actually very difficult.  What I mean by this is that when I use Toggl, any distraction -- to read Reddit, eat food, go to the bathroom, no matter how brief -- is counted as something other than work.  This means that I regularly spend ten hours a day in the office, but only get about 7 hours of actual work done.  ...If someone is coming in at 9am and leaving by 5pm, they're certainly not working an actual 40 hour week.  Run a timer for yourself, and you'll be amazed by what I mean.

At my last three month review, I was just starting as a Software Engineer Intern, and had not done any actual work yet.  In the past three months, I have successfully completed my internship, been hired as a permanent Software Engineer, and then transferred to become a Data Scientist.  In doing so, I learned a lot more about computer programming and a lot more about data science.

I think my job has been the most valuable for the immense amount of learning.  With my day job and another ~9 hours a week of additional programming training, I think it's not an exaggeration to suggest that my programming knowledge has more than doubled, even though I only went from 700hrs of programming work to ~1100hrs of programming work.  Having mentorship is pretty useful.  I strengthened my knowledge of Ruby on Rails and R, and added knowledge in Coffeescript, Angular, Knockout, Mongo, Express, Node.js, VimScript, and Shell script.

Also, my day job has afforded me the ability to donate more than $3K to date, and save even more than that to donate later.



As an off-shoot to learning how to program, I've started working on a few "passive income" entrepreneurship projects -- the kinds of projects that I hope will bring in some additional money, but not require too much maintenance.  I think doing projects like these is good for (a) building programming skills, (b) building management/entrepreneurship skills that may be valuable if I want to work on start-ups later, (c) building career capital, and (d) potentially earning more money.

So I compiled a list of ideas and got to 54 of them.  Many of them sucked, but I got to work on a few of them that I thought might not suck that much.

The first project I launched was, which is a website that tells you what type of processor your computer is (64-bit or 32-bit).  The website contains an ad, which has made ~$8 to date.  Not quite enough to cover the $10 of domain costs or the ~5 hours of programming time, but still a good start in that I have officially made money online and learned a bit.

I have a couple more ideas for ad-supported websites that I want to launch that might go live in the future, but ultimately I'd like to get away from that and toward other monetization models.  I hope by the next review I'll be announcing the launch of some much meatier ideas that are currently in the works, and more than $8 in revenue.

But currently, like all great companies, I am in the red.

Direct EA Work

One of my main career goals since I started thinking about careers has always been to keep a careful eye toward whether I'd be more valuable doing direct work.

So far, I've experimented with 13 different EA projects.  I hope to pare this down significantly to the top performers soon, as I don't have that much time to spend on these projects.

  • Charity Science: Charity Science is the EA org being run by Joey Savoie, Xio Kikauka, and Tom Ash that aims to fundraise for GiveWell top charities.  So far they've raised a good bit of money.  I've been helping them out with strategy.
  • The Birthday for Charity Project: One of Charity Science's fundraising campaigns has been to encourage EAs to raise money for their birthdays.  This has been a surprisingly good value for a very small amount of time.  All my work on this project sums to 2.5 hours, and the project has brought in nearly $9K.  Split that among four people, and it's still $900/hr. I don't think I've ever had returns that good before.
  • EA Survey Data Analysis: Back in May, .impact launched a survey of effective altruists (still open if you want to participate).  I'm working on doing the data analysis for that.  I'm almost finished.  Expect results by the end of the month.
  • A Study of Facebook Veg Ads: I'm working to study the effectiveness of vegetarian advocacy Facebook ads on getting people to go vegetarian.  Work is slow, but steady.
  • EA Profiles and the EA Donation Registry: Recently launched were EA Profiles and the EA Donation Registry.  I helped a bit with putting that together, but it was nearly entirely the work of Tom Ash.
  • This EA Forum: I helped a bit with advising Ryan Carey on setting it up and being part of the initial seed content.  I hope to write more blogs here to keep my karma count competitive. ;)
  • EA Job Board: The EA Job Board exists to help catalog jobs related to effective altruism.  It still needs a few more features, but it's mostly finished and completely usable.
  • 80K Trial Research: I've been working with 80K on a trial basis to do some career research.  I did a bit more research into foundations.  Still need to work a lot more on this.
  • Chicago Skeptics Meetup: Charity Science has had a fair amount of success by networking with atheists and humanists in Vancouver.  I went to the Chicago Skeptics Meetup to see if I could do the same thing.  I only have been able to go once, so results are inconclusive, but I found another EA there and convinced him to sign up with Giving What We Can, so it didn't go all that badly.  I'm interested in going more, if just for social reasons.
  • Animal Rights Protesting: Direct Action Everywhere came to Chicago, so I went out to try out animal rights protesting and see what it's all about.  It's exceedingly difficult to figure out if my participation had any value, but it was fun and a good way to expand my own comfort zone.  More notes on this are forthcoming.

I'm still not very clear on my value as a EA direct worker, relative to my value as a programmer / entrepreneur who can move money to direct workers.  I think there has been a ton of value in having very high engagement with EA orgs that I donate with, however, and direct work has really helped with this.


In my previous review, I spent a lot of time moving again, switching to my second apartment.  I think this was worth it, as I cut my rent down by $255/month (Chicago is expensive) and cut my commute down by 20 minutes a day (valued at $400/month) for a place I consider just as good (better in some aspects, worse in others).  I expect it to be more than three months or more before I move again, so the 30 hours spent will ultimately get me ~$65/hour in value.  Though moving is pretty stressful and I don't want to do it too often.


Sleep, Break, and Social

Sleep is 8 hours a day, which I consider pretty typical.  I'm considering experimenting with different methods of sleeping less (e.g., naps, caffeine, maaaybe modafinil), but ultimately I'm pretty concerned that the desire to sleep less is a fallacy and actually I do more on 8 hours of sleep, since I'm well-rested enough to do good work.

Time spent on break is higher than in the last review, but I see this mostly as an artifact of switching to Toggl, and therefore being able to penalize the short 2-10 minute breaks that I used to not count.  While I think it would be more productive to cut this down, I think I'm at a pretty ideal point in terms of my break amount and spacing.  While I don't want to break to excess, they are pretty important for restoring focus.  Like sleep, I think breaks may be underrated by people who are focused on productivity.

Lastly, my social time continues to be quite high at 27 hours a week.  While down from the 37 hours a week of my summer time, it's not the focused 15 hours a week I achieved in February.  I'm excited, however, to make social connections here in Chicago, and think spending a large amount of time upfront making connections is pretty justified.  A lot of my social time is also spent with my girlfriend, who I like a lot.  Furthermore, I went on vacation for a week, went to Gen Con, and went to a college reunion, all three of which made social time larger than what would be normal.


Prepping, Planning, Email, Food, Cleaning, Errands, and Exercise

It's nice to have Toggl now because my estimates for these activities will be much more active.  There's not a whole lot I have to comment on any of these.  Some people may be interested in how a lot of these times are so low, and I plan to offer more discussion about that at a later date.  (Though the reason why exercise is so low is because I've fallen off the exercise wagon a couple times, and I hope to have this be a larger amount by the next review.)



I've been very excited about the rich variety of activities I've been able to partake in, from socializing to my day job to a rich life of EA and entrepreneurship projects.  However, I've also felt the cost of having my attention divided in too many places, and not being able to have enough time to focus on any individual thing to the degree that it deserves to really see it through.

I've been experimenting with restructuring my schedule to find more time to do things, and have had a fair amount of success with this in the past week.  However, I think I'll soon be faced with the harsh reality that I'll just need to prioritize things and decide which things are providing the highest amount of value, and ruthlessly end the remainder.  I don't currently know yet which remainder to end, though.

Ultimately, I'm still interested in whether I should continue my day job, or whether I should pursue independent work (either in entrepreneurship or in direct work), though I feel a lot less conflicted in this than I was a month ago.  I think that the day job is a good direction for me in terms of providing real skill building, good career capital, status, and the ability to donate.

Three months ago, I was just moving into Chicago, having no idea what is ahead of me.  In many ways, things are different than they were in college, where I knew what I would be doing a year ahead.  Now, I'm not even sure where I'll be -- both physically and in terms of my beliefs about what my future should be -- by my next review.

I think the wisdom of keeping options open is very wise, and by my variety of activities, I think I've done this to it's fullest extent.  I've often found that I have no idea what value certain projects will have until I do them, and so -- until then -- I'll just continue to do.

It's amazing how much you can get done if you just keep on doing things.

14 comments, sorted by Click to highlight new comments since: Today at 6:53 PM
New Comment

Back in May, .impact launched a survey of effective altruists (still open if you want to participate). I'm working on doing the data analysis for that. I'm almost finished. Except results by the end of the month.

Glad to hear it! I look forward to seeing the results.

This is very useful. As someone still very new to this who wants to contribute more it can be helpful to see what other EAs are doing in detail. I still struggle from not knowing exactly what I can do now and what are realistic goals for behavioral and social changes, particularly in the short term.

More generally, as someone trying to be more productive and efficient Toggl looks promising and I'm going to try it out myself.

Thanks for the feedback. I think Toggl is worthwhile, but as a warning it's pretty hard to get into. I started my time tracking via Google Calendar first (which is easier) for 15 months before getting into Toggl. So don't be too dismayed if it doesn't work out. But I think the knowledge of self is definitely worth it, even if you track for just a week!

Do you have any recommendations on how to avoid wasting time updating the current activity on Toggl?

1.) Don't have that many Toggl categories, so you don't have to update that often.

2.) Use the green button to re-start a project rather than typing it over again.

3.) Don't stare at the timer.

The standard advice is just to try it for a week or two. It can be useful to become more aware of how you're spending your time. Doing it permanently is a bit hardcore :)

Lots of people track their work hours over the long-term, but it's usually just through a spreadsheet or gcal.

Interesting review as always Peter. I've looked at Toggl's tour (and like its name for reasons you can claim a prize for guessing next time we chat ;), but could you say a bit more about what made you pick it over alternatives? I ask as someone who's done time tracking for a while, and settled on but isn't totally satisfied by it (mainly because of the time-consumingness of tagging my time in a fine-grained way).

Paul Christiano and Katja Grace jointly explored several different time-tracking apps and concluded that, while Toggl has flaws, it has fewer flaws than the alternatives. Independently, I did some research myself a while ago and also came to the conclusion that Toggl was superior to its rivals, though I can no longer recall the considerations that led me to reach this conclusion.

It would be cool if Paul and Katja published their results.

Also, I remembered the reason why I use Toggl is because you recommended it to me.

Though, like with Paul and Katja's recommendation of Workflowy, I was skeptical at first.

I prefer Toggl because it is fastest: a single keypress to stop the current timer or return to the last one, and a single key press followed by typing a uniquely-identifying substring to start a new timer. Katja and/or me also tried yast, freckle, clok, harvest, lumina, and a few other options I've forgotten about. All of these have more clicks/keystrokes per operation.

Toggl also works offline, which ruled out several alternatives, and has a reasonable (though not excellent) interface for viewing reports, and 3 levels of reporting (name of entry, job, and client).

Note that Toggl crashes constantly (perhaps 2-4 times a day for me on OSX), but it can be restarted in a few seconds each time, and so this isn't a huge consideration for me.

Note that Toggl crashes constantly (perhaps 2-4 times a day for me on OSX), but it can be restarted in a few seconds each time, and so this isn't a huge consideration for me.

Huh. Is that the desktop app? I've never used that app, but have had 0 problems with Toggl in any of it's other manifestations (website, mobile, command line tools).

I didn't look at alternatives, so I can't say why I picked it over alternatives. Toggl has a nice interface. Paymo looks fine, though!

(Typo: "Except results by the end of the month" -- should be "expect", right?)