Post-food living — a quantified experiment

The social studies, sensing experiences and well beings of living entirely on powder and water for a month

Like many others, I lived on Soylent and water exclusively for a month. While doing so, I logged the experience to be without food.

First off, two things. I lost 6 kilograms during the experiment, mostly fat. If that is all you want to see, then yes, powder food for dieting works. If you are interested in the many other aspects of this extreme diet, please read on.

Secondly, I lied (a little). I never tried Soylent, being European. Instead I used the Dutch equivalent called Joylent which uses the same recipe but adds (in my case) a subtle vanilla flavour.

Rules for living food-less — overview of experiment scope and methodology

I planned the experiment to test how living on powder food would impact me as a person, both in terms of health, physical well-being and activity, but importantly also how it would impact my social life, my relationships and my productivity. To test all this I set up the following rules:

  • Breakfast has to be whey protein powder mixed with water (to get 40 grams of protein in the morning)
  • Lunch and dinner has to be Joylent (1/3 of daily recommended for each meal, meaning circa 166 grams/meal)
  • In between, only water and tea are allowed as intake (tea is the British/Indian/Chinese way, with no added sugars or anything, mostly whole-leafed types)
  • No other intakes
  • No stimulants, food based or not (part of the experiment is being sober to better absorb the mood aspects of food)
  • Logging 5 times every day, before, during and after the trial
  • Exercise is allowed, but not especially encouraged or mandatory (to reflect the same exercise level as in before experiment)

To quantify my experience I set up a system for tracking different physical and mental elements multiple times a day. These are explained more in their respective chapters, but are:

  • Sleep
  • Mood
  • Appetite
  • Exercise
  • Sugar intake
  • Work
  • Alcohol intake
  • Eating (other than powder food)
  • Stool quality

But without further explaining, let us delve into the results of this experiment.


I am a little late for work, the one day of the week where this is not acceptable. I don’t care that much. I am not looking forward to it, the mask of insomnia making itself apparent on my face, making the thought of looking into a computer monitor filled with spreadsheet numbers exhausting already before I reach the building. I showered colder this morning, just trying to squeeze a little awaken energy out of myself. I remember every hour of the night before, having had no sleep at all. I have tried this before. More so I am afraid of it happening for a whole week again.
11th of November, 3 days after ending powder food trial

I have always had problems with sleep. 5 hours is a good night for me. A year before the trials I went a week without sleeping at all, without knowing having any clear idea of why. I have since then focused a lot on what could improve my sleep, hoping for general better life quality thereof. Underneath is my sleeping quality during the testing period. Important to note is the sleeping quality is noted based on my opinion in the morning, not quantitative sleeping data.

Time series of sleep quality during experiment with powder food versus baseline periods. 42 data points, 23 (55%) on powder food, 19 (45%) off powder food

Generally, it seems the sleeping quality stabilises once the body has gotten any transitions periods. Interestingly, the worst sleep during the powder food trials were just after (transition period) and just before transitioning back off powder food. The latter however, have no apparent relationship with nutrition. Percentage values for the data is shown underneath.

Percentage distribution of sleep data during experiment. 42 data points, 23 (55%) on powder food, 19 (45%) off powder food

So, I sleep better when I am only eating powder food. But there are factors alone that scew the non-powder food results. Other than nutritional variables in the intake (especially sugar in my case), alcohol is a large contributing factor to better and worse sleep. To show this, the chart underneath compares the quantified averages of different sleep measurements during the trial period.

Average sleep quality for different nutrition/alcohol series. Data points after colour: Green 42, red 23, light blue 19, darker blue 5, purple 14


I have always been pretty stable in my moods. But I have also been acutely aware that sugar might have the same effect on me as a drug would, with highs and lows, peaks and valleys.

The theory behind this is that having no added sugars and other stuff in my food would stabilise my mood.

For tracking my mood I used a model called the circumplex mood model. The model does not allowing for ranges between emotions, but functions well for quantifying specific emotions.

Circumplex mood model (

I assigned each mood on an xy graph, from -4 to +4, skipping zero. This lead to the graph underneath, the size of bubbles showing the frequency of the mood during the testing period.

Moods and their frequencies during test period. Data points on powder 126 (61%), not on powder food 80 (39%)

To make the data a little clearer, I sorted the data into the bar chart underneath with percentage values.

Percentage distribution on mood. Data points on powder food 126, not on powder food 80

The graph shows that I am a dominantly content person, for better or worse, with or without powder food. I was more bored while being on powder food, for reasons explained later in the article. I think the slight larger share of relaxed mood might be due to the sobriety of the month.


I logged my appetite throughout the experiment to see if the different method of eating would change my hunger throughout the experiment. With powder food, there quickly seems no point in holding three meals per day, instead opting for always having some around for consumption whenever the need arrives.

Percentage distribution for hunger. Data points on powder food 126 (61%), off powder food 80 (39%)

Surprisingly, despite not distributing my daily intake in three meals, instead taking a more fluid (pun intended) approach to my intake.

It is not surprising that I felt lesser instances of being full. However, that I had more instances of slight hunger is a surprise, and honestly a disappointment. Having the ability to consume food at any time should erradicate the feeling of hunger very quickly, but apparently did not.

Bowel movements

This is getting a little personal, I know. But stool is the easiest way to know the general wellbeing of your gut. The method is pretty reliant too.

By using the medical standard of Bristol Stool Form Scale, where you look or feel (during passing) the form of the stool, the gut health can be easily assessed.

Bristol Stool Scale (

I don’t think I have a particular weak stomach, but certain kinds of food have an effect on me, especially the unhealthy kind. I was a little excited to what the diet would make my stomach feel like.

Percentage distribution of bowel movements. Data points on powder food 11 (37%), off powder food 19 (63%)

How you eat have a large impact on how your stomach acts and how often. To get another perspective on my gut health, I logged when I had bowel movements.

Time series of bowel movements. Data points on powder food 11 (37%), off powder food 19 (63%)

It is noticeable that the frequency of bowel movements varies greatly, both with the different data series and between the series. It shows that there were irregularities just after each transition. This seems fairly natural after a complete diet change.

The data also shows a farther space between the data points during the powder food experiment than during the reference periods. During the reference period, the average bowel movements per day was 0.94, meaning usually once per day. While I was eating (drinking) powder food, the average was 0.41, more than two days between. Powder food, ingesting only 500 grams per day with no excess calories, leaves a minimal amount to extrude.

The unwritten side

There were things during the experiment I discovered, but not through data points. The practicalities of my life became much easier. I only went shopping twice in 30 days, none for food or drinks. It was really easy building habits around not having to take time out to eat.

At the same time I felt become more secluded, like interactions were less important in a sudden. I stopped taking time out for lunch at work, since I didn’t use for eating anyways. That meant I missed the majority of social interactions with my collegues, even though I only started working there a month prior.

I knew my family, especially my parents disapproved of my experiment. I knew that if I went home to them, they would likely try to bully me into eating something other than powder food, try to make me brake my experiment. I knew this because every phone conversation had some hint of it. I stayed away from visiting my parents and generally became even more distanced than usually, directly avoiding direct contact with them, not wanting to take the discussion.

I slowly stopped being social, since all interactions with friends involved food and/or drinking alcohol. At the end of the experiment I couldn’t think of who to call to celebrate, so I ate my first meal in a month alone. It was about as sad as it sounds.

Two weeks before ending the experiment, I did a short film project where I had to drink fake champagne, made of tonic water and elderflower (which tastes fantastic, btw). After each taste, I spit out the drink, not wanting to ruin my experiment even a little. I became just as extreme as I needed to be.

I went on a second date with a girl during the experiment. I scrambled to find a date not involving food or alcohol. It ended up being bouldering and tea, pretty good though. She later on found someone else. I blame some of it on my developing disinterest in social interactions. I never saw her in person after the date, even though I liked her.

In the end I became a little worried that the self-inflicted seclusion I was under would become something like a depression, even though I am not prone to this.

At the same time, I felt I had more mental energy than usual. Since I ignored social stimuli, I turned to difficult working and learning tasks. I started e-courses, spend time learning some useful programming and developing projects. It was one of my most productive and creative periods in a long time.

Physical improvements and problems

I had taken a Bioelectrical Impedance Analysis (BIA) before and after the experiment. In 39 days, 28 of them on powder food diet, I lost 6.1 kg (13.4 pounds) and my body fat percentage went from 23.3% to 20.1%, pulling me away from the edge of medical obesity.

Change in weight and BFP during the experiment

Of the 6.1 kg weight loss, 4.3 was pure fat, 1 kg being muscles and the rest being water. I had been very careful about not exercising more than before the experiment started, even though my rising energy level dared me to.

At the end of the experiment, I developed annoying hiccups that wouldn't easily go away, unless I drank water in a certain way. I am convinced that it was due to under extercised throat muscles. The hiccups were gone less than two days after I started eating solid food again.

The conclusion

Switching to consuming only powder food definitely is both for the good and bad. My social life took a hit, but I felt years younger at the end (pretty impressive, considering my 24 years of age at the time). It showed me the power different habits has on your physical and mental wellbeing.

My mood where data correlates to have eating non-powder food. 34 data points

I still eat powder, quite often. Almost every meal that I eat because I’m just hungry, I make powder meals. I stress eating experiences more than before my experiment too. I try to optimise the nutrition fulfillment of my life more. I have a bag of powder food and a shaker in my office draw, in case I work late. Every evening I make a small shake of protein powder for my breakfast, because I like to just get up and leave out the door as fast as possible.

My mood where data correlates to drinking alcohol. 11 data points

A thing I do now, is that I keep regular schedules of social activities with people. Every Thursday I hang out with one friend, every third Wednesday with other friends, at Saturday I always have something planned. It’s important to do things that make you happy. As seen on the graphs at the end, for me it isn’t eating, but sometimes drinking. I never drink alone, which explains it.

If you have any questions, write me on Twitter. If you found it useful for you, bookmark it. If you thought useful for others, please recommend,

One clap, two clap, three clap, forty?

By clapping more or less, you can signal to us which stories really stand out.