I drank 1000 cups of coffee; here is what I learned ☕️

“Processed data is information. Processed information is knowledge Processed knowledge is Wisdom.”

I love tracking data about myself.
Recently I heard of the term “self-tracker”, so I researched about it and ever since, I adopted the term and happy to describe myself as one.

The concept of using technology to collect data about myself and then analyzing that data just out of curiosity, or to better understand different aspects of myself has always sounded interesting to me, both as an engineer and as a researcher.

These personal analytics I collect range from tracking how much coffee I drink every day to a list of all the movies I ever watched. And without any doubt the obvious things like the number of steps I walk every day or number of hours I sleep at night, just for me to have a better perception of my behavior patterns.

Of course, the process of self-tracking is not the goal. The magic happens when the collected data is put into context and I’m reflecting on what it means. The goal is to use your data to answer your questions.

The main question I asked myself about my coffee consumption was

“how many cups of coffee I drink every day?”

I had a pretty good assumption of the number, but as a good researcher, I had to back it up with real data.

First coffee cup

My journey began on 11.10.2015; I recorded the first coffee cup. 📅
The beginning was pretty hard as I tried to stay motivated and keep logging every cup until it became a habit. Every time I drink a cup of coffee, I record it.

The collection of data can be done with the help of technology, but it is not necessary. You can track yourself however you want, even with a notebook.
To help me track my coffee consumption, I use my “Samsung Gear S2 classic” watch that I wear almost 24/7 ⌚️and the “Samsung Health” app with the built-in coffee tracking feature.

Then I occasionally export Samsung health data as a CSV file and use my coffee intake analytics open-source project to extract the insights from it. (You are welcome to contribute your lines of code to improve it)

Check out the project here:

dimshik100/Coffee-intake-stats-extractor

The data recorded about every cup from the Samsung health app is pretty straightforward and contains the time and the date, and some other less relevant fields. Here is an example record from the CSV file:

So after tracking my coffee consumption for about three years, and 1000 cups later, I did find out how many cups I drink every day.

Source: Wikipedia

Just kidding 😂…

The simple average is

(number of cups) / (number of days) = average cups per day

Some days I drink a lot of coffee

I found out that it depends on what day of the week I’m looking at. For example, on Saturday I rarely drink coffee, but on Sunday I drink the most number of cups.

Besides that, I found many additional interesting things. What is the shortest time between me drinking two cups of coffee?

Or the maximal number of cups drank in one day.

I even built a coffee analytics single-page application that uses the extracted data to tell if I’m drinking a cup right now (well, It presents the probability percentage of me drinking a cup right now). I plan to add there some more cool statistics I extracted from the data, like a Pie chart of the percentage of cups distributed by the days of the week.

Check out the app here:

Is DiMsHiK Drinking coffee right now ?

One big idea I had was to combine the dates and times of me drinking coffee with my geolocation data collected by Google. With this information, I can create a prediction machine that will show the probability of me drinking coffee in a specific location. Something like

“Dimshik is going to drink coffee in the next hour in Aroma Tel Aviv”

This way, you can use it to meet me for a coffee without ever scheduling with me, which can be very cool or very creepy.

By the way, this is me drinking yet another cup of coffee.

And this is a very cool coffee I drank a while ago that was brewed using a Japanese siphon.

Let’s summarize:

1. Collect data
2. Process it to get information
3. Process it to get knowledge
4. Process it to get wisdom
5. Create crazy and fun projects
6. Have fun :)

Call to action

Source: art4clip

This is an open-source project, so feel free to suggest more interesting insights that you think can be extracted from this data set, Or even contribute to the source code, Improve functionality and performance.
Please leave your comments below or through my Twitter.

This article is a part of the Self-tracking series.

Read more:

1. How to extract your personal Samsung Health data
2. I drank 1000 cups of coffee, Here is what I learned (You are here)
3. I saw 1500 movies. Here is what I learned