Minding the Borderlands

Mark Koester (@markwkoester) on the art of travel and technology

A Matter of Fecal Matter: A 31-Day Scatological Self-Tracking Experiment

I tracked my poop for a month. Here is how I did it, what I did to track and process the and, in the end, what I learned from a matter of fecal matter.

First off, I did not literally touch or photograph my poop during this experiment. What I did do is log each and every time I passed a stool.

There are some few alternative ways to track your poop. Interestingly, there is a lot of talk in health and self-tracker space around the “gut microbiome” testing. This is where you analyze the bacterial makeup of your fecal matter, and there are several commercial companies that offer this service. The New York Times seems to love writing about it (here is a good article to get you started here). One of my favorite podcasts, The Quantified Body, has several in-depth episodes on the microbiome too. . It’s a topic that merits a separate experiment and discussion.

Similarly, in case you didn’t know it, there are already quite a few apps dedicated to tracking and logging your excrement. Apps like Poo Keeper, Poop Tracker and others, let you log and rate your poop using the Bristol Stool Scale (BSS). Developed over 20 years ago, this 7-type poop categorization system has become the gold standard for the clinical evaluation of your poo. This short medium post offers a great intro into the Bristol Stool Scale and poop tracking.

For this self-tracking experiment, I decided to keep things simple. I used a generic tracking tool called Hindsight to keep a log of my body waste over a 31-day period. Basically, each time I pooped, I took a few extra seconds to log the activity. Unfortunately I have yet to find a passive way to track my poop (yet).

While there are probably better ways to spend my time and plenty of other more “appropriate” experiments you can do to quantify your life or track your health, poop tracking provided an interesting and amusing opportunity to test out this new lifelogging tool, to practice my skills in data analysis and data visualization, and to learn a few more things about myself and my poop.

In this post, we will be looking at lifelogging and data visualization of fecal excrement over a month-long period, using Hindsight app.

Why I Tracked My Poop

Before getting into the “how,” I have to answer the obvious question about why I tracked my poop.

There are a lot of reasons people track and quantify their lives. In their Five-Factor Framework of Self-Tracking Motivations, researchers Henner Gimpel and Marcia Nißen identified several reasons behind why people track their lives:

  • self-healing, i.e. people aim at becoming healthier
  • self-discipline, i.e. people do it for the rewarding aspects of it
  • self-design, i.e. people like tracking because of the control and ability to optimize “yourself”
  • self-association, i.e. self-trackers and quantified self enthusiasts are associated with movement, group or community
  • self-entertainment, i.e. people track because like other hobbies, there is an entertainment value

I track different aspects of my life for different reasons, and I find this list of motivations provides a good way of framing the why of my own tracking.

When it comes to fitness tracking like running or mobility, I do it for the self-discipline that tracking it provides. I can look at my training logs and calendar schedule as reinforcement for improving my fitness performance. I track my tasks and time, because it helps me design a particular kind of life and gives me a level of control and understanding that wouldn’t be possible if I didn’t track my productivity. I track and study several aspects of health, because I believe that tracking my health provides a way to establish a healthier baseline, prevent future problems, and live longer. As such, forms of health tracking like blood testing and monitoring my heart rate, blood pressure and heart rate variability act as a form of self-healing.

Strange as it might seem, when it comes why I tracked my poop, it was primarily a form of entertainment for me. Like watching movies or reading books, I track my poop because it was fun to do.

What is Lifelogging?

Lifelogging is considered to be one of the more extreme sides of self-tracking and quantified self movement. Back in 1994, one of the original practitioners Canadian Researcher Steve Mann continually broadcasted his life 24-hours a day online using a wearable computer. Several similar projects have followed as well as a Microsoft wearable camera project that allowed users to create a regular stream of first-person photos.

In a past post, we’ve looked at the idea of lifelogging and some tools you can use to capture life’s miscellaneous like Reporter App and and my personal favorite, Nomie.

Lifelogging typically involves the idea of creating a comprehensive lifestream of data and media from one’s life. Historically it meant recording your life in live streaming video. But in fact life logging has a much simpler meaning too, and I like to consider lifelogging as a form of tracking where you “tally,” count or notice different things in your life. For example, this could be how many coffees you drink or how often you smile, complain, or hug.

One of my favorite examples of lifelogging in this sense of noticing and visualizing comes from Giorgi Lupi and Stefanie Posavec, who spent a year exchanging postcards filled with different data they collected. Each week, the two picked some area to collect data on, like “thank yous,” and productivity and then shared hand-drawn postcards that expressed that data in some way. They have since published a book on it called Dear Data and “Note to Self” podcast share their tale in Facing Our Weirdest Selves.

What I like about this form of lifelogging is that, while you are tracking and noting something, it is more broadly about noticing. Life logging makes it possible to put concrete examples onto some area of your life you hadn’t thought about before. I would argue that the value of life logging doesn’t come just from the data, but from the noticing and reflection from that noticing.

Hindsight: A Lifelogging App to Track “How Long Since…?”

Over the past month of poop tracking, I used a simple yet effective self-tracking tool called Hindsight. Hindsight is an iOS and Apple Watch app that lets you log when anything happened in the past and then provides a timer telling you how long since that event occurred.

This tool in effect provides a kind of support memory since lets know when was the last time you called your mom, got a haircut, trimmed the lawn, or any other common activity we tend to forget when we did it last.

One of the key advantages of this tool over other lifelogging and tally tracking apps I’ve used in the past is that I can log things from my Apple Watch without my phone. Increasingly I consider the Apple Watch as the ultimate self-tracking tool, and Hindsight is a great example of accomplishing something on your watch without the potential distraction from doing it on your phone.

How I Tracked My Poop and Initial Data Exploration with Google Sheets

To track my poop, I create an activity type in Hindsight, and each time I went to the bathroom and pooped, I logged it there.

Hindsight app does not provide detail reports or charts, like I get with Streaks, my favorite habit tracking app. It does a great job for simple data logging, providing a simple key metric (time since X…) and all of the logged data is all exportable.

After exporting my data, here is what the initial data looked like:

While this data looks decent, it took me a bit of work to dig into the data and process it in a more usage format.

Like most data projects, it’s important to imagine what your end goal is. In my case, I wanted to know how often I pooped per day, generally around what time I most often pooped and what was the typical time delta between poops. With these goals in mind, let’s look at how I used Google Sheets to help with the initial data processing.

First, I wanted to group poop events by time of day. So, I extracted the hour of the poop using this formula:

=HOUR(FIELD-REFERENCE)

This effectively allowed me to have one-hour bins in which each event occurred. Putting data into hourly or minute bins is a good technique when working with time-series data.

Second, to calculate the delta between events, I used a coupe of clever google sheets tweaks. Here are the steps I took:

  1. Create a new field called Time
  2. Reference the value of occurrence field with =C2
  3. Change the format to “Time”
  4. Update all the existing data with this formula
  5. Then create a new column called DeltaCalc
  6. In DeltaCalc, a simple formula that subtracts the difference between current and previous field like this: =E3-E2
  7. Copy the results from DeltaCalc and paste as values only into a new field
  8. Convert Delta Results to Seconds with this formula: =value(H3*24*3600)
  9. Convert Seconds Results to Hours with this formula: =I3/60/60

These steps allowed me to calculate the time delta between events and provided the key metrics I wanted, the time delta between poops.

Following steps 1 and 2, here is what the data collected like:

Third, I wanted to know how many times I pooped each day. This was a bit more cumbersome to determine, but here is how I did it:

  1. Create a new field for date.
  2. Add this Formula =DATEVALUE(FIELD) that extracts date from the time stamp of the event.
  3. Copy the results of this calculated field
  4. Paste into that column as “Values Only.” This will remove the full date and time and give you a simple grouping reference.
  5. Create a pivot table with “Date” as Row and Date as Value with COUNTA.

The end result is a breakdown of how many days you pooped once or more.

Finally with all the results in Google Sheets, it’s easy to create a simple pivot table with average, max, min, median and standard deviation:

Visualizing a Life In Pooping with Tableau

Now that we have cleaned up our data, created the necessary values and done some initial data exploration in Google Sheets, it’s time to visualize it. Here is the end result I created using Tableau:

Interactive Graphic @ https://public.tableau.com/profile/mark.koester#!/vizhome/PoopTracker-AScatologicalSelf-TrackingHistory/PoopTracker

Conclusion: Observations and Lessons from Tracking My Poop

In this post, we looked at poop tracking using Hindsight App. In the first part, I shared a bit about the history of lifelogging, some tools you can use to “tally” track nearly anything, and more about the specific tool I used, Hindsight. In the second part, we moved into how I organized this tracking experiment. Finally, in part three, we looked at the logs of data I collected and how to do a bit of data processing with Google Sheets, before creating the final data visualizations using Tableau.

In the conclusion, I’d like to provide some of the lessons I learned about myself and my poop as well as some ideas for future experiments.

The most high level “learnings” this experiment provided were these:

  • I tend to poo twice a day (61% of days).
  • I most frequently poop in mornings with 74.5% of my poo-ing occurring before 11am. (While not cross-validated, the most obvious culprit being “coffee makes you poop.”)
  • Nearly all of bowel movements occurred within 24-hours of the previous one, excluding two exceptions: once while traveling and another during a 36-hour water fast.
  • Nearly 50% of my bowel movements occurred between 18 and 24 hours of the last.

In short, I poop in the morning, most often after drinking coffee. I didn’t need a month of data logging to tell me this, but it’s interesting to see the patterns in the data.

Interestingly, this tracking experiment coincided with a few unique events: a 36-hour fast, running a marathon and some international traveling. Marathons require early morning wake ups and it’s best to run after passing a stool. So that’s one of the reasons for some early morning variance.

One new thing I did learn was that fasting, i.e. intentionally not eating for a period, makes me stop pooping, even when I drink coffee. I’ve been fasting off and one for a few years, but I have only recently combined this with additional health tracking.

The tail end of this experiment concurred with the kickoff of an on-going daily monitoring of my glucose levels. So, these additional metrics provide some interesting background material to better understand my bowel movement patterns too, though I didn’t see anything noteworthy yet.

For now, I do not plan to continue logging my poop beyond this 31-day “scatological” experiment. I have a baseline of my excrement behavior if I ever need to compare when I change living situations, foods I eat, get sick, etc. Like blood testing, I don’t necessary thing it’s valuable by itself, but having this baseline in the future might be useful if and when I get sick or have a health change.

There are two additional poop related experiments I’m keen to run. The first would be a bowel transit time test. Everybody’s “bowel transit time” varies. So, if you want to know yours, the suggestion from gastroenterologists is to determine yourself with what’s known as the “corn test.” Typically done using corn, the goal of this experiment is to know how long it takes a person to go from digestion to excretion. Eat some corn and then watch for when the undigested kernels show up.

The second one would be a gut microbiome test. As mentioned in the intro, the microbiome has become one of the most discussed areas in personal health and tracking. Basically our bodies are filled with different bacteria, and using basic genetic analysis and sophisticated data science, researchers are starting to learn more and more about how the balance of different bacteria affect our health. For example, one well-known example relates to how certain antibiotics lead to the loss of much of the diversity in our gut microbiome and some negative health problems. This area remains quite new and there is even variance between different testing companies due to how the tests are done and their reference data sets. Like blood tests, a microbiome test can provide a very interesting baseline for future reference too.

Hope you found this post as entertaining reading as I did running and writing about the experiment. Leave a comment or send me a note with your thoughts!

Comments