Tracking a research protocol means turning a string of one-off actions into a record you can reason about. The value is not the logging itself — it is what a complete log lets you see that memory simply cannot.
In plain terms: write it down as you go, and the useful numbers fall out of the record for free.
What to record
A log is only as good as what it captures. Five fields carry almost all the value:
| Field | What it is | Why it matters |
|---|---|---|
| Compound | What was taken | Makes every later comparison meaningful |
| Amount | How much | The input to the level curve |
| Time | When each dose happened | Timing drives accumulation and decay |
| Site | Where an injection went | Enables rotation |
| Inventory | Vial contents and doses left | Predicts run-out and age-out |
Anything you measure over time — a lab marker, a body metric — is a sixth, optional layer that lets you look for trends.
Why logging beats memory
Across every self-tracking domain the finding is the same: recall is unreliable and selective, while a record made at the time is neither. Medication-adherence research is explicit about this — self-report gathered after the fact is prone to recall bias, and the longer the gap between the event and the report, the worse the accuracy1. Contemporaneous diaries are recommended precisely because they sidestep that memory drift2.
In plain terms: your brain rounds off and reorders repeated events. A written log does not.
For a protocol, an accurate record is what lets you answer "when was the last dose," "how many are left," and "what does the level curve look like right now" — none of which memory answers well.
From log to insight
A raw log is just data. The useful part is *derived* from it:
- Run-out dates — from dose size and how often you log.
- Level curves — from dose timing and the compound's half-life.
- Rotation suggestions — from your history of injection sites.
- Adherence — from logged doses measured against a planned schedule.
None of these require extra work if the underlying log is complete. That is the whole argument for logging carefully once, rather than reconstructing later.
Zyra Labs is built around exactly this: log the dose in a few taps, and it derives the schedule, the inventory depletion, the estimated levels, and the site rotation — locally, on your device.
The honest limit
Tracking tells you what *happened* and what your numbers *are*; it does not tell you what a protocol *should* be. That is a clinical question. A log is a mirror, not an advisor — and everything on this page is a method for keeping a clean record, not a recommendation to run any protocol.
The short version
Record what, how much, when, where, and how much is left — as it happens. Do that, and the calculations you actually care about come for free. Define the schedule first so the log has something to measure against.