I started tracking time 4 years ago. It was new and exciting in the beginning. After a little while, I found it tedious and exhaustive.
Eventually, it became a habit.
When I started, I didn't know what to track and how to do it. Should I log time only for projects at work or for anything? Should I track how much time I spend watching TV?
I didn't have these answers. Most importantly, I didn't have any goals neither –– I just wanted to try some kind of lifelogging because the idea was quite attractive at that time.
You can't be disappointed if you don't have any expectations, so I was positively surprised by the results I was able to achieve:
- Less context switching
- A realistic picture of the effort
- Better estimation
Less context switching
When I say "time tracking", I mean "manual time tracking".
I open an app on my computer or smartphone, fill a couple of fields about what I'm doing and press "Start". Or I just select something from the history if I did it in the past.
Nobody complained, but I suppose that sometimes it can be annoying for people around me.
Imagine the situation: I work on something on the laptop, and a colleague comes with a question
Colleague: Hi, do you have a couple of minutes?
Me: <picks the phone>
Me: <opens an app, does several taps>
Me: <puts the phone in place>
Me: Sure, what's up?
Luckily they know that I'm not just checking my Instagram feed –– but it may look a bit weird. Just a bit.
The reasoning behind these strange movements is that I want to have accurate statistics. More or less. When someone asks me for a couple of minutes, it never takes 2 minutes. It seems clear that I should stop my current timer –– otherwise, data would be incorrect.
Well, you say, but what's about online communications? If you track when somebody distracts you offline, shouldn't you do the same when you get a new message?
Of course, I should. And I do! When something online requires me to stop doing my current task and switch my attention, I log it as well.
And this is the main idea – I deliberately add extra overhead for a context switch.
It is normal for people to avoid extra work, right? The solution is simple: if I need to do all this extra work when I switch my attention, I unintentionally try to reduce context switch.
It's like procrastination back to front. When you have a hard task to do, your brain wants to switch to something simple.
The concept is the same: it's harder to switch to the new thing because you need to track time. Therefore it's easier to continue your current task.
You don't look for distractions, you avoid it.
A realistic picture of the effort
I had this feeling many times in my life: I try hard, but nothing changes.
- I work on my project, but it never becomes new Facebook. Actually, it never even goes public.
- I try to grow as a professional, but I still don’t feel like I’m ready to apply for this position.
- I work 12 hours a day but don’t get a promotion (my favourite)
Whenever I hear it, I advise tracking time to get the real picture.
Usually, it's properly explained by this tweet:
but it's less funny.
Data in this research illustrate that people overestimate their work hours by 5–10 per cent and the more hours they say, the bigger the gap is:
And this is only about "working hours" –– how much time was spend on work. We come to the office every workday at the same time (more or less) and still give the wrong answer.
Imagine how it goes if we should answer about "productive hours" or our own projects, where we don't have any schedule and guides:
- I feel like I learn Spanish, but the data says that I spent on it only 2 hours last month
- I feel like I’m exhausted working on my project, but in reality, I have been programming for only 30 minutes. And then I spent 3 hours reading articles about Elon Musk. Yes, probably it was somehow related, but it doesn't help my project.
If you feel that you work hard, but something is unjustly bad, try to get the realistic picture of effort.
This part is a bit different from the previous two.
You track time –– you get less context switch.
You track time –– you get a picture of your effort.
But it doesn't work like this with estimation. It's sad to admit but to be better in estimation you should not only track time, but also analyse data.
Good estimation requires us to solve 2 problems:
- Estimate how much time a task can take
- Estimate when we finish
Yes, these problems are different. If I need 8 hours to finish this task it doesn't mean that it'll be done tomorrow. But first things first.
To estimate a task, I usually:
- Decompose it to small chunks
- Go through all the sub-tasks and try to remember if I did something similar in the past.
If I did, I roughly know how much time it took for me to finish (because I tracked it).
If I didn't have anything like this, I try to guess.
Then, after I finish the task, I look at my log and my prediction to get more information for the future. Next time I have something similar, my estimation will be more accurate.
To estimate when we finish, we need to know how much time we have. There are a lot of things that can use our productive time, such as meetings, context switch, breaks and other unpredictable activities.
And this is the place where time tracking can help a lot. The more data we have, the more we know how much time we spend on different activities. Yes, it'll be different from day to day, but the average value can give enough information.
For example, last year I spent 40% of my working time on actual tasks. If I estimate, that a task will take 20 hours or more, it's highly possible that I won't finish it in one week. Therefore if someone depends on it, I'll warn the stakeholders right away.
Probably you can find more benefits –– this is just how it works for me.
I tried a lot of different strategies for the last 4 years, but this is too much for one note. I plan to write more about this topic in the future, and I'll be happy to know about your experience and answer your questions.
Feel free to reach out to me on Twitter or somewhere else