An experiment is a short, bounded "try it and measure it" loop you run on yourself. You change one behaviour, NOYU watches one metric, and at the end you get a plain-language read on what your own numbers did.
It is the difference between reading that morning light helps sleep and knowing whether it helps your sleep.
The shape of one
Every experiment has four parts, all decided before you start:
- The change. One behaviour, and only one. No caffeine after 3pm, get outside within an hour of waking, no alcohol for two weeks.
- The metric. One thing NOYU measures to see whether it moved. Your sleep score, your HRV, your sleeping heart rate, or something you rate yourself like energy or mood.
- The window. From a fortnight up to six weeks, set to match how long the effect would plausibly take to show. A sleep-onset change can read in a fortnight. A fitness change needs longer.
- The reason. A short explanation of the biological mechanism, with the research it rests on.
Stating all four up front is what keeps the result honest. If you pick the metric after seeing the data, you will always find something.
Why the window length varies
Different changes surface on different timescales. Cutting caffeine earlier can show in how quickly you fall asleep within days. Raising your easy cardio volume works through your cardiovascular system over weeks, so its experiment runs for six.
NOYU sets a sensible default per experiment and you can adjust it. A window that is too short is worse than no experiment, because it produces a confident-looking answer built on noise.
One at a time
You can run one experiment at a time.
This is deliberate, and it is the whole point rather than a limitation we are working around. If you change three things at once and your sleep improves, you have learnt nothing about which one did it. One change is the only version of this that can answer a question.
Your journal, your scores and everything else carry on in the background regardless. The single-experiment rule applies only to experiments.
What NOYU will and will not claim
An experiment is not a clinical trial. There is no control group, nothing is blinded, and life keeps happening around it. NOYU is direct about this rather than dressing it up.
So the result tells you what your own metrics did across the window, and names the things that could reasonably explain it other than the change you made. It never tells you the change is proven. It is signal, not proof.
Where an experiment has research behind it, NOYU shows you the strength of that evidence and links the study. Only well-supported changes make it into the library at all.
What is not in the library
Every experiment in the library is a behavioural change: something you do, or stop doing.
There is nothing you swallow. Supplements and other ingested substances need safety vetting we have not done, so we have left them out rather than shipped them loosely.
