The short answer: a useful peptide log records where each vial came from, not just when you used it. That means the compound, the vendor, the lot or batch number, the COA for that exact batch, the dates it arrived, was opened and was mixed, how it was stored, what you noticed, and when you had bloodwork.
Most tracker apps are built around the calendar: what, when, which site. That part is easy. The part almost nobody records is the vial's paper trail, and it's the only part that helps when something goes wrong.
Here's what to write down, and why each line earns its place.
Last checked October 21, 2026. This is about record-keeping, not dosing. Nothing here is medical advice.
Where I get mine: the vendors I use, and how I vetted them.
Why keep a log at all?
Because when something goes wrong, nobody else has the details. Not the vendor, not your doctor, and not the researchers trying to work out what these products do.
Look at what happened when researchers studied what people report about unapproved retatrutide. They read 148,640 Reddit posts and comments from 38,936 users. The limitation they flag, in their own words: "Product identity, purity, dose, route, treatment duration, concomitant medications, and medical history cannot be verified" (JAMA Network Open, Oct 1, 2026).
That sentence describes most people's records too. A good log fixes about half of it. You can't verify purity from your kitchen, but you can know exactly which batch you used, what its COA said, when you mixed it and what you noticed.
What should a peptide tracker record?
Ten fields, in two groups: the vial's paper trail and your own notes.

Field | Why log it | Example entry |
|---|---|---|
Compound | The obvious one. Write the name exactly as it's printed on the label | "BPC-157, as printed on label" |
Vendor | You can't follow up on a problem with a seller you can't name | "Ordered from [vendor], order # on receipt" |
Lot or batch number | Ties this vial to one production run, its COA and anyone else's report on it | "Lot: as printed on vial (photo saved)" |
COA for that batch | Proves what the batch tested as, and lets you check the lot matches | "COA PDF saved, lot matches vial, lab name noted" |
Date received | Starts the clock on storage and shows how it arrived | "Arrived Oct 3, cold pack melted" |
Date opened and date mixed | Mixed vials age. Most have no published stability data at all | "Mixed Oct 5, bacteriostatic water" |
Storage | Where it lived and anything that went wrong | "Fridge, back shelf. Left out overnight Oct 12" |
Dose (when and where) | Timing and site matter for spotting patterns. Keep this to what your own plan says | "Dose taken, morning, left abdomen" |
Observations and side effects | The only record of what you noticed, in your words, on the day | "Redness at site, gone in two hours" |
Bloodwork dates | Lets you line up lab results with what you were using at the time | "Panel drawn Oct 20, results in labs folder" |
Two habits make the table work. Take a photo of every vial label the day it arrives, so the lot number survives a smudged label. And save the COA file next to the entry, not as a link, because vendor pages change.
I use PeptIQ for this: peptiq.io. The parts that match this list are vial tracking (the App Store listing says it tracks "vials on hand, amount left, low stock and expiration") and bloodwork, which it describes as "Upload bloodwork and keep lab results in one timeline" (App Store listing). Each entry takes "date, time, site, and notes," and it exports to CSV or PDF (peptiq.io/features). Whatever you use, the fields above are the point, not the app.
Why do lot numbers and COAs matter?
Because a COA only means something for the batch it tested. A lot number is how you prove your vial came from that batch.
A COA, or certificate of analysis, is a lab report on one production run. If the lot on your vial doesn't match the lot on the COA, the report isn't about your vial. If you've never checked, here's how to read a peptide COA and what the numbers on it mean.
Lot numbers matter even in the regulated world. In December 2025, FDA warned about counterfeit Ozempic labeled "with lot number - PAR1229 - which also is an authentic lot number" (FDA, Dec 5, 2025). A lot number alone doesn't prove a product is real. But without one written down, you can't even check.
Two more reasons:
Batches differ. Australian researchers tested three products labeled with the same retatrutide amount. All contained retatrutide, but "One product contained only half as much as stated on the label," and the other two had more than claimed (JAMA, Oct 9, 2026). If you notice a change partway through, the lot number tells you whether you switched batches.
Testing needs it. If you send a vial to an independent lab, the result is only useful tied to a lot. More on that in how to test peptides for purity.
Why record the date you mixed a vial?
Because the clock that matters starts when you mix it, and for most research peptides nobody has published how long that clock runs.
There's no published stability data for BPC-157, TB-500, CJC-1295, ipamorelin, MOTS-c or retatrutide once they're mixed with bacteriostatic water. The "28 days" rule you've seen is a sterility rule for multi-dose vials, not proof the peptide is still intact. I went through what the labels and the little data that exists actually say in how long mixed peptides last in the fridge.
What a mix date gives you is a way to connect a change to the vial's age. If something felt different in week four of a vial and not week one, you'll only know if the date is written down. Same for storage slips. "Left out overnight" is a one-line note that could explain a lot later.
Why log side effects, and where do you report them?
Because a side effect you didn't write down didn't happen, as far as anyone else will ever know. And reports on unregulated products are still thin.
JAMA put it plainly in October: "There have been scattered reports of adverse events associated with black market retatrutide." The same piece cites a Public Citizen count, based on FDA adverse event data, that "at least 14 users have been hospitalized," and six cases of liver toxicity reported to health officials in Victoria, Australia, linked to products labeled retatrutide (JAMA, Oct 9, 2026).
Two more data points from the same month:
A published case. Doctors in Zurich described a woman in her mid-20s with five days of nausea and vomiting that didn't respond to standard treatment. Only a detailed history turned up weekly injections of "an unverified product reported to contain retatrutide." The authors' advice to other doctors: ask explicitly about "injections, peptides, research chemicals, online products" (BMJ Case Reports, Oct 7, 2026). Your log is how you answer that question in thirty seconds.
Poison center calls. America's Poison Centers logged "862 exposures to retatrutide in the three months ending in September," up from 460 the quarter before. CBS also reported two deaths submitted to FDA, while noting "the details aren't verified and it is unclear whether the drug was to blame" (CBS News, Oct 13, 2026).
None of that says your vial is dangerous. It says the record is thin, and the records that exist are hard to trace back to a product. A log with lot numbers is the fix.
Where to report. FDA's MedWatch program takes reports from "health professionals, patients and consumers," not just doctors (FDA MedWatch). A report with the compound, vendor, lot number and dates is far more useful than one without. If a reaction is serious, get care first. The log can wait.
How to write the note itself:
Write it the same day. Memory rewrites symptoms fast.
Describe, don't diagnose. "Itchy red patch, coin-sized, at site" beats "allergic reaction."
Note when it started and stopped. Timing is most of what a doctor will ask.
Log the good days too. "Nothing unusual" is data. It shows the bad day was different.
Why put bloodwork dates in the log?
Because lab results without context are just numbers. With a date, you can see what you were using when the blood was drawn.
You don't need the full results in the log. You need the date, the lab, and where the report lives. Then, months later, you can line up a change on a panel with the vial, the lot and the week. My own first month on retatrutide is written up here if you want an example of lining things up by date.
Paper, spreadsheet or app?
Any of them works if you fill in the fields. The format is the least important decision here.
Format | Good at | Weak at |
|---|---|---|
Paper notebook | Fast, no setup, nothing to sync | Can't search it, easy to lose, can't attach a COA |
Spreadsheet | Searchable, sortable, free | Clunky on a phone, so entries get skipped |
Tracker app | Quick to log on the day, reminders, photos and files in one place | Only as good as the fields it has. Check it records lot and batch, not just doses |
The test for any of them: can you answer "which lot was I using on this date, and what did its COA say?" in under a minute? If yes, your log works.
The short version
Log the vial's paper trail: compound, vendor, lot number, the COA for that lot, and the dates it arrived, was opened and was mixed.
Log your side of it: storage slips, observations and side effects in plain words, and bloodwork dates.
A COA only covers its own batch. The lot number is how you prove the match.
Mixed vials age, and most have no published stability data, so the mix date matters.
Side effects from unregulated products are under-reported and hard to trace. MedWatch takes consumer reports, and a lot number makes yours useful.
Leave dosing math out of your log's job. This is a paper trail, not a protocol.
Reply and tell me what you log that I didn't list here. If you've ever traced a bad reaction back to a specific batch, I especially want to hear how you did it.
