For a two-week period recently (April 27 to May 11), I time-audited myself. That is, I recorded everything that I spent time on, in 5-minute increments. I’m happy to share with you what I learned from both the resulting data and the process.
I started this project somewhat spontaneously. I found myself frustrated for a lack of time, and wondered, “Where does all of my time go, after all? Why don’t I try to measure it?”
This is a project that I found very rewarding, as it helped me gauge exactly what I spend time on. I didn’t see it as a moral judgment, but obviously there are some natural suggestions for improvement.
When he heard about this project, one of my friends asked a simple question: “If there are things you don’t want to do, why don’t you just stop doing them?” My response is that everything that I did over the course of these two weeks, I wanted to do, either because I enjoyed it or as means to another end.
The main purpose of studying my time like this is to try to gauge where the different tradeoffs are. What am I going to have to spend less time on if I want to do more of X?
Unlike most scientific papers, you’re going to have to keep reading to get my conclusions. 🙂
To record the raw data, I used a Google spreadsheet that looked something like this:
|Figure 1: The spreadsheet. Note: I later renamed the “Research” category to “Math.”
I simply listed every category of activity I took part in, and added up in 5-minute increments how much time I spent on each in a given day. The “Updated To” row helped me as I retraced what I had done since the last update, and I highlighted the row of the current day so I could find it easily. (Since my typical sleep schedule is around 1am to 9am, I decided to put my transition between days at 5am.)
Where I was doing multiple things at once, e.g. walking and reading Feedly, I just listed it as the one where my attention was based. If I toggled between activities, I tried to estimate how much time I spent on each the next time I checked in.
Surprisingly, this was not a huge burden on me, either time-wise or psychologically. I also had a slot for this recording, and in all, I estimated I took 165 minutes doing the recording over the course of two weeks.
In fact, after the first week, I was so enthusiastic about it that I wanted to do it all the time. At the end of the second week of recording, though, I was starting to wear down, so I decided it was a good time to stop.
On a positive note, this practice did make me much more aware of where my time was going. Recording everything led to some moments where I would ask myself, “Where did the last 15 minutes go?”, which helped me notice when time was just slipping by.
I categorized my time in a couple of different ways.
First, I picked out some major categories of my life, and plotted the time spent on each:
|Figure 2: The macro-categories of my time. All units are in hours per week.
Most of these should be pretty self-explanatory, but I might as well describe them.
- Health obviously includes sleeping, as well as exercising and showering.
- Math is primarily research and a reading seminar I was a part of.
- Social is the broadest category, including my relationship with Grace, talking with friends in my department and dorm, and keeping up with old friends and family.
- Christian primarily includes my involvement in my church, community group, Graduate Christian Fellowship.
- Special is an unusually large category these two weeks, covering my community group’s camping trip and time spent hosting a friend on my futon.
- Fun consists of Facebook and other reading-type distractions, keeping up with Major League Baseball, and games nights.
- Tasks includes all sorts of miscellaneous tasks, the biggest of which was reading my e-mails.
- Writing is this blog.
- My Teaching involvement these two weeks was all with Art of Problem Solving.
The second way I categorized my time was by “tiers.”
|Figure 3: Tier Classification of time. All units are hours per week.
You can see the brief descriptions, but here are the top things in each tier (bold = more than 4 hours/week):
- Non-negotiables: Sleep, e-mail, cleanliness, walking, food, laundry.
- Top Priorities: Grace, exercise, research, bible study/prayer, church.
- Chosen Involvement: reading group, blog posts, AoPS, church community group, Graduate Christian Fellowship (GCF) large group, GCF large group planning, my dorm, this time audit.
- Good Ideas: vacation (the camping trip), keeping in touch, meals with friends, hosting visitors, concerts/ceremonies, seminars.
- Fun Extras: Facebook / internet, baseball, games, Feedly.
Finally, I’m not bothering to compare the first week to the second week, or do any sort of time series analysis like that. Instead, I’m taking this to be a two-week snapshot of my life.
In the introduction, I hinted that I wanted to spend more time on something, and by now, it should be obvious that I need to spend more time on research. I averaged less than 20 hours a week doing any math-related work, and that is simply far too low. I’d like to increase that by another 20 hours at a minimum.
But before getting to where all of that time is going, let me share some positive observations:
- My lifestyle is pretty healthy. I averaged 54 hr/wk (7.7 hr/night) sleeping and over 5 hr/wk playing frisbee (not including sports played while camping).
- Spiritually, my habits of daily bible study and prayer are strong (20 min/day average).
- I spend a lot of time talking to Grace (over 1 hr/day), and as a result our relationship is in a good place.
- I’m able to stay in contact with many friends, and managed to plan a decent number of meals with them (2.8 hr/wk).
- Baseball season is in full swing, but the total time I spend on it (2.7 hr/wk) amounts to less than one game length per week.
- Despite it counting as non-negotiable, I only spent 1.1 hr/wk on food, thanks to nearly all of my meals having another purpose.
Now, where is the fat to trim from all of this? I’m estimating these differences relative to another typical week during the year.
- Most obviously in the snapshot above, I spent almost an hour per day on Facebook and other internet reading (6.8 hr/wk). That is a lot. This time isn’t all “wasted” — many friends post interesting articles on Facebook that I like to read. Some even give me ideas for future blog entries. But an hour per day is a lot, and I would do better to reduce that to around 2 hours per week. Let’s say I get 5 hr/wk out of this.
- The Special category was unusually high at 13.6 hr/wk. The camping trip features prominently, and it’s tempting to just say that I can typically get those 9 hr/wk back. Here’s how I’ll justify it: I probably had enough special occasions these two weeks to span around six typical weeks, so I should be able to save about 2/3 of that time in a typical week.
- My official responsibilities in my dorm (Sidney-Pacific) are ending now; the 2.4 hr/wk here included my last house meeting and a transition brunch to the new government. Let’s say I bring that down by 2 hr/wk.
- Of course, I won’t be auditing my time every week, so I get 2 hr/wk back from that.
- I attended more seminars than usual in this time (with the Simons Lectures in the math department), so say I get back 1 hr/wk out of the 2.6 hr/wk I spent this week.
- As I find more people to delegate to, planning GCF large groups will reduce from about 2.7 hr/wk by around 1 hr/wk.
Collectively, those get me my required 20 extra hours to spend on research. Notably, all but the Facebook hours are sort of “automatic”; I don’t have to think about not spending time on my dorm since it’s out of my hands now.
Of course, the natural problem is that just as I whack these moles down, some other things are going to go up in time. For instance, without going camping, I wouldn’t miss church Sunday morning, so that’s another two hours (1 hr/wk) that comes back. My games group was also basically inactive during that time, so I’ll probably add another 3 hr/wk when that comes back.
Where is all of my time going, then? If I probably can’t reliably spend even 40 hours per week on research, where does it all go? Here’s Tier 3 again:
- Reading group: 10.0
- Blog: 4.4
- AoPS teaching: 4.4
- Church Community Group: 4.2
- GCF Large Group: 4.0
- Large Group Planning: 2.7 -> 1.7
- Sidney-Pacific: 2.4 -> 0.4
- Time Audit: 2.0 -> 0
- Misc Christian Responsibilities: 1.2
- AD: The Bible Continues: 1.1
- GCF Prayer: 0.3
After the reading group (which counted as research-related), the heavy hitters are the next four, which apart from this blog are my Wednesday, Thursday, and Friday evening activities. And I’m pretty committed to all four of them. It’s the middle three that I’m seeking to reduce.
And here’s everything in Tier 4 with more than 1 hr/wk:
- Vacation/retreat: 9.1 -> 3.1
- Keeping in touch: 3.6
- Meals with friends: 2.8
- Hosting visitors: 2.8 -> 0.8
- Concerts/ceremonies: 2.8 -> 0.8
- Seminars: 2.6 -> 1.6
- Department socialization: 1.5
- Remainder (mainly misc tasks): 2.5
Of course, the main fat to be trimmed here comes from that special section. The rest of this category consists mainly of friendships I actively try to maintain.
Here’s a simpler way to look at it, on a per-day basis. I’ll start with my daily activities:
- Tier 1: 10 hours (sleep, clean, e-mail)
- Tier 2: 2 hours (Grace, exercise, bible study)
- Tier 4: 1 hour (catching up, meals with friends)
- Tier 5: 1 hour (Facebook, baseball)
After that, I have 10 hours per day left, for a maximum of 70 hours to spend on work and/or activities. If I spend 6 hours per day on research to reach 42 hours per week, that leaves four hours remaining per day. Here’s what fills those four hours:
- Monday: Blog-writing (ahem, ideally)
- Tuesday: Large Group Planning and other tasks
- Wednesday: Community Group
- Thursday: Art of Problem Solving
- Friday: GCF Large Group
- Saturday: Special occasions
- Sunday: Church, AD: The Bible Continues
Of course, these estimates are on average; I can easily spend more time on research on Tuesdays and spread the tasks throughout the week.
Finally, I do need to think about how to avoid being on Facebook for an hour per day, getting that down to around 20 minutes per day. I’ve previously tried to use the Chrome extension StayFocusd to limit my time, but too easily found ways around it. A decent portion of the time I spend on Facebook is on my phone, for instance.
What’s different now? Of course, the moral force of this public commitment will count for something, I think. In addition, I also noticed that I would often spend this time mindlessly browsing the web when I was relatively brain-dead, i.e. in the morning before I went in (or got out of bed), or late at night.
The week since then has suggested some solutions to me. With activities starting before 10am four of the last five days, I’ve seen the benefits to having something in the morning to go to. I’ve also tried to get myself to work in the mornings, but too often fall asleep while reading a paper. My next step will be to not read a paper, but instead work at a white board in the mornings, and make sure to go to bed as soon as I finish my to-do list at night.
I set out to measure my expenditures of time in order to try to find ways to spend more time doing math, trying to increase my typical 20 hours per week to around double that amount. Given some of the particulars of the two-week period which I measured, I was able to find that time, mainly by natural decreases but also calling for a substantial reduction of around 5 hours per week in the time I spend on Facebook and other internet distractions.
This exercise has mainly increased my literacy and understanding of how much time various activities will cost, and where that time will come from. My continued time commitments in terms of hours per week cluster like this:
- Large (~8 hr/wk): Relationship with Grace, keeping up friendships
- Medium (~4 hr/wk): GCF Large Group, church Community Group, Art of Problem Solving teaching, this blog, exercise
- Small (~2 hr/wk): Bible study / prayer, Large Group planning, seminars, department socializing
I do plan to do this again some time, to collect more data and measure what’s changed. I’d also be curious if anyone else tries it, to compare our results.