Sometimes when I’m hanging out with group of friends, fellow writers, or co-workers, the topic of quantified self comes up. Someone will say, “Ask Jamie, he can tell you how many steps he walked from here to the meeting room. Or how many words he wrote in the last 15 minutes.” Everyone will laugh, the word “nerd” will be used affectionately, and the conversation will drift on.
Inevitable, someone will approach me afterward and say, “I think it’s really cool that you capture all that data about yourself. I want to know how to do it, too.” Sometimes people worry that they have to know how to write code. (Not necessary, but it can’t hurt!) More often, they worry that it will take too much effort.
Over time, I have evolved a set of guidelines to identify what to track, what to avoid, and how to do so with as little effort as possible. They are:
1. Start with data that is captured automatically.
2. Start with data that doesn’t require a judgment call.
3. Start with data that is easily accessible.
Start with data that is captured automatically.
Most of us are busy enough without adding yet another item to our to-do lists each day. If you want to start capturing data about yourself and your behavior, I suggest starting with data that is captured automatically.
Activity-trackers are a good example of this. You wear a FitBit Flex, Jawbone Up, or Nike Fuel Band on your wrist, and go about your normal day. The device captures the data for you. There are apps you can install on your smartphone, like the Moves app, which tracks your activity. I have scripts that automatically track how much I write each day.
These devices automate the process of collecting data as you go through your day. While you walk, your activity tracker captures data. While you drive, another device or app captures data. You don’t have to remember to turn something on or off. That is good because having to take an extra action reduces the likelihood that you’ll continue to track the activity.
Start with data that doesn’t require a judgment call
Sometimes, data collection requires a judgment call—this happens with data that isn’t collected automatically. There are many applications that can track the food you eat. Aside from not being automatic—you have to take an additional action beyond eating your meals to record them—they require you to guess at things like portion size. Your guess may be good, but it is still a judgment call.
Another example is mood tracking. Mood tracking is very popular in the quantified self arena, but I have avoided it because (a) I have to remember to do it; and (b) it requires a judgment call on my part. Maybe everyone else thinks I’m in a foul mood, but I think I’m perfectly happy.
This type of data is valuable, but I give it a lower priority because it requires extra work to collect. This doesn’t mean you shouldn’t collect it, but if you’re getting started, I’d recommend prioritizing it below data that is collected automatically and requires no judgment call about the data.
Start with data that is easily accessible
There is no point in collecting data if that data is not easily accessible for you to use. Oftentimes, you can look at your data in the app that collects it. I can see my FitBit data in the FitBit app, but I also want access to the raw data so that I can play with it and learn from it.
When looking at what to track, especially when starting out, I recommend keeping two questions in mind:
1. Is the data available in an easy-to-use format? Can it be exported to Excel, or a comma-separated file, for instance?
2. Does it cost anything to access the data? Some services charge a fee for premium functionality, and sometimes, this functionality includes exporting the data.
For those with coding experience, you might also consider whether the data is available through an API. Once you have ready access to the data, that is where the real fun begins. It is also where the real learning begins.
These guidelines are intended to provide a framework, something to keep in mind when getting started. For some people, tracking food intake or mood may be an important personal goal. These guidelines should help you understand the level of effort involved in capturing the data, so that you can make better-informed decisions.