I currently log the total number of hours I work each day in an OpenOffice Spreadsheet.  I input the start time, lunch/break time, and end time, and it calculates the total hours worked.  I'm not savvy enough to create this type of spreadsheet myself, so I looked through a large number of templates online before finding one that works as I've described above.  I'm still not crazy about the way that this spreadsheet is laid out.  

If you can link to a spreadsheet available for download similar to the one described above, please do so in the comments.  

If you use time-logging for various distinct projects throughout the day, please describe this process and link to the software you use (if possible). 

More of a meta-discussion: how time-logging this enhanced your performance or time management?, for what types of projects/activities is it best to time-log?, general comments about the idea

New Comment
20 comments, sorted by Click to highlight new comments since: Today at 9:03 PM

Usually I think it is to be a headache for deploying spreadsheets, linking it to other stuffs. Rather I prefer using an automated time tracking software from Replicon ( http://www.replicon.com/time-tracking-softwares.aspx ) which works with a click and ultimately the the automated process gets inclined with.

As I note below, RescueTime is an excellent app for this. It's very fine grained, automatically keeps track of every program you use and website you visit separately down to the second, and requires basically no interaction to run except to check up on yourself at the end of the day; if you tell it which activities/websites are productive it'll also tell you how much of the time you're actually working, and how that gets distributed about the hour/day/week/month. (It probably goes up to year as well, but I haven't been using it long enough to check.) Overall I am very impressed with it.

At MIRI we use Harvest to track employee and contractor time, and it's been largely successful.

We track our time against different projects and have one General project that we use when we aren't working on something that makes sense to create a separate project for. One of the reasons we decided on Harvest is because their reporting tools are really great, which is important to us. The other one is that even though it's fairly powerful, the interface is easy to work with.

My preferred way to use their system is with timers. When I start working each day I start a timer on our General project and process my inbox and figure out what I'm going to work on for that day. After that, I start timers as needed for the various projects I end up working on. With their apps it's really easy to switch from one timer to another and add a new timer so I haven't found doing so to burdensome. Timer are also available through their web app which is nice.

Their exporting functionality is quite robust as well so you don't have to worry about having your data locked in.

I'd also have a look at Freckle. They were my number one choice in terms of user experience, but they lacked a feature we needed (though I can't remember what it is anymore). Looks like they have made a lot of improvements since last I checked.

I wrote a simple text parser which understands semi-free text like



$P_1$ +30m reading lesswrong

$A_1$ +1h project A: implementing feature 1

includes discussion with Pete


-15m lunch break

+15m dicussion about new feature 2

This allows me to take notes with extremely low overhead. I just switch to my programmers editor and write down a few more lines. No opening window, not clicking or tabbing thru redundant fields.

And I can do so incrementally. The $P_1: token assigns the cost center and I usually assign these at some later time batching related items quickly and possibly creating the cost centers after the fact - which is usual at the beginning of a project. The cost token is looked up via a comparable file describing the hierarchical structure of cost centers.

When I started this it was intended to just track time for myself, but it developed into a tool which can push the data into Jira (via REST) and Kimai (via SQL) and export selected sub hierarchies as TeX reports for invoices.

But it is still a simplistic undocumented CLI tool run from my IDE with lots of idiosyncrasies, so I wouldn't recommend it for anybody else.

I believe that it (i.e. the use of plaintext) is a lot more efficient and powerful than a spreadsheet.

A bunch of Quantified Self people use Google Forms. I think it's probably an improvement about OpenOffice spreadsheets.

I've tested several programs for time logging and todolist is the most versatile program I've found. It's free and also has a portable version.

You can log tasks, subtasks, create time estimates and compare actual time spent and estimate entire project times and project completition in percentages based on subtask completition. You can schedule and prioritize. You can tweak options to make the program just as simple or complex as you like.

I'm not sure it can be used to export spreadsheets or why you would want to do that.

Sounds (and looks like) a Windows-only version of Task Coach, which is my favored tool.

I'm still unclear on whether tracking time this way does usefully improve my estimates of the amount of time I use on tasks. It does seem to increase my motivation to finish tasks (and then move on and finish another..), though.

Thanks, wasn't aware of the program. I tried it and it seems a lot simpler and less modifiable, but that might be a good thing depending on how you look at it.

Nitpicks: It seems to open a new window every time you create something, that definately slows things down. There seems to be few hotkeys. In ToDoList there's a customizable keyboard shorcut for everything and I rely on them heavily.

There's about .. 24 or 30 shortcuts, yeah (mostly shown in the rightclick menu). In practice I find these match the core actions I do (new task, new subtask, edit task, start/stop time tracking, add notes, mark as completed..).

The new window thing bugs me too (but probably for different reasons, as a designer I think it's the correct choice and they should have just made it faster, probably by using GTK+ instead of Qt)

I have been using this app for about 6 months:

Pomodoro Calendar

I have found it to be very, very beneficial in terms of my productivity and focus*. It's based on the pomodoro technique, which follows the basic pattern of working in 25 minute chunks. Whenever you do a work chunk, it can be automatically synced with google calendar. At the end of the day, it's easy to see how many work chunks you have done. Additionally, it can be set up so that you can easily create preset categories for the type of work you are doing. I have categories for different projects at work, hobbies, and exercise. Before I set the timer, I simply select which category I'm doing my work in, and then hit go. When I look at my google calendar, I have an honest breakdown of how I'm spending my productive time.

A word of warning - the app is a little rough around the edges. Every now and then (twice a week?), it will crash, and syncing with google calendar requires a little troubleshooting to start. This wouldn't be that difficult for anyone on LW, and once it's set up, syncing takes care of itself. Crashing is an insignificant issue, and at worst means you don't record the work done during that chunk.

*I now measure my productivity (or at least my work ethic) in how many chunks I have done in a day. I do substantially more now than when I first started, which could be partly because I have gotten into the habit of measuring myself. Still, I feel that I am actually working more. If anyone has any suggestions on how I could more fairly compare my productivity now vs. in the past, I'd be happy to give it a try.

I often hear rationalists seeking out things like this, but I've yet to hear any of them outright assert or even imply that this is a useful thing to do. I myself have thought about such things before, and my answer has been that I switch tabs and contexts too rapidly to accurately measure these things. In order to get anywhere near effective measurements, I'd need interfaces magnitude orders more pluggable than existing software implementations allow. (Firefox, for example does not value programmability or pluggability, but rather extensibility.) Despite the usefulness of such projects, they will take considerable effort, and I've simply not been able to motivate myself to bother simply for the sake of tracking time spent. I won't argue it doesn't have any benefits, even known benefits, but I have yet to see any kind of evidence that it is legitimately useful. Perhaps the evidence is not found in the places I've expected it to be thus far.

If anyone can justify the benefits or give me tangible evidence of any such thing, I would appreciate it.

I have basically this problem (I'm fairly sure it's tied into my ADHD). As noted below, RescueTime is an excellent solution: it'll keep track of different tabs in the same program separately, do it automatically, and do it down to the second - so you get a very accurate result at the end. I've found some very valuable results from this - for starters, that I work only about 50% of my "work hours" even on a good day, and that I spend much more time on random surfing than I thought I did.

As a baseline, I need a program that will give me more information than simply being slightly more aware of my actions does. I want something that will give me surprising information I wouldn't have noticed otherwise. This is necessarily non-trivial, especially given my knack for metacognition.

By definition, I can't really guarantee that information it gives you will be surprising. I can tell you that I consider myself a fairly luminous person and that RescueTime still managed to surprise me.

At any rate: I also don't think you're going to get better than RescueTime. It keeps track of everything, does it down to the second, and does it without you having to notice - and it certainly helped me, so there's one data point.

I see. I'll have to look into it some time.

If we track "hours worked" only, that metric might be of little use toward productivity. The first thing that comes to mind is to track hours worked against goals accomplished / milestones met, evaluate how effectively we use our time to achieve various things, and then work on improving inefficiency.

If we track "hours worked" only, that metric might be of little use toward productivity.

Depends on the given person's level. Where I am now, the main problem is completely avoiding the work, instead wasting all my time online. If I succeed to put at least 1 hour of work towards my goals every day, then I am ready for the next lesson, which is making this hour more efficient.

Meta: Perhaps we could start by describing various levels of goal-reaching-inefficiency, creating a self-diagnostic questionnaire, and then give specific advice for each level. Something like:
Level 0 = you are not even able to log whether you work or not. Advice: unknown.
Level 1 = you are not working. Advice: use Pomodoro.
Level 2 = you are working only on low-priority stuff. Advice: prioritize.
Level 3 = you are working on short-term high-priority stuff, but you don't have a strategy for increasing your skills. Advice: set apart some time for long-term improvement. Etc.

I see. So it works as an estimation in determining which things to prioritize effort into optimizing.

I track my time using RescueTime. The value to me is improving my calibration with respect to how well I feel I'm working, as compared to my actual RescueTime hours. Sometimes I think "Wow, I really worked a lot today" when in fact I didn't get many hours in, and I'd rather have my intuition match the metrics. I don't have a special justification/goal beyond that but I'm hoping something useful pops out.

I suspect that this is an instance of low cost, low median outcomes, but with high upside -- it's unlikely you'll find something that makes a difference, but the cost isn't very high, and there's always a chance that, without putting numbers on it, you are missing some productivity intervention which would make a big difference to you. For example, perhaps you think poorly when you're sleep deprived, but you don't know it, but tracking productivity would let you know that's happening.

At the NY Quantified Self meetup a few weeks ago, somebody reported tracking her post-concussion symptoms and discovered that, in fact, she wasn't suffering from a concussion at all -- it was a very different condition which required separate treatment.

Ah, I see. That clarifies things significantly. It also further indicates my erratic context switching is non-ordinary and requires special needs on this front.

New to LessWrong?