I'm comparing two sources' public COAs and hit a weird pattern. One report shows HPLC purity 98.7%, lab date Jan 12, but the batch code on the vial photo is AX-2207 while the COA says AX-2209. The other source's COA has a QR that just opens the lab homepage.
I've spent two weeks building a small checklist—lab name, report ID, sample date, QR destination, batch match—and I'm stuck on what actually proves a report belongs to a batch. What signals do you treat as disqualifying?
Batch mismatch is often sloppy relabeling, not proof of fake. I'd ask for the original lab email header, not just a PDF.
QR to hompeage = decorative. I ignore it.
Decorative, sure. I email the lab directly with report ID and sample date. Some confirm, some cite privacy.
That's the catch. I look for a verification portal instead. If the source won't walk me through it live on a call, I pass.
The portal point is why I did a six-week check of 14 public source pages—not purchases, just what they show before contact. I logged COA date, lab, report ID, batch code, QR destination, and whether the lab's own portal recognized the ID.
Nine had batch mismatches or missing IDs, three never answered a verification email, and only two had a working portal. The portal ones still had one report where the sample date was later than the COA issue date. That inconsistency mattered more than purity numbers. I know that's a small sample size.
The date thing is easy to miss. I now sort by sample date first, then purity. If the sample date is after the report date, it's an automatic no.
Sorting by sample date helps, but paper-only labs are my sticking point. I've seen scans with mismatched fonts, but that's a hunch, not data.
Honestly the QR-to-homepage thing is the least suspicious part of what you described — I've pulled COAs from labs that are legit and their QR just dumps you on the homepage, so I stopped weighing that one entirely. The mismatch that would actually stop me is the batch code, and here's a trick I picked up after getting burned: I ask for a photo of the vial label sitting next to a handwritten slip with my order number and that day's date, because a repackager can slap any sticker on a vial but it's a lot harder to fake that in one shot.
The other tell is the timeline — I keep a spreadsheet of order date, batch code, COA report ID and sample date, and the one time something was off, the sample date on the COA was about nine months older than the batch it claimed to be, which meant they were just recycling an old report. So my question back to you: does the Jan 12 sample date line up with when that batch would plausibly have been made, or is it older than the batch code suggests?
That gap has caught more junk for me than any batch number typo, and weirdly the spreadsheet habit is the same one that's kept me honest on my own tracking since I started, down 18 lbs and still logging like a nerd.
My only real batch mismatch turned out to be my own handwriting—AX-2207 vs AX-2209—so now the rule is: ask for the original PDF instead of a screenshot, check the creation/modified date, search the report ID on the lab's own portal, and line up the sample date against the date the source actually shipped. If the file was created after shipping, or the ID comes back empty, the purity number can be as pretty as it wants, it doesn't matter.
The other thing that saved me from panic-switching sources was a dumb little Sunday habit: waist at the navel, before coffee. Down 2.5 inches over 9 weeks while the scale only moved about 6 lb. Which is a decent reminder that a weird COA doesn't automatically mean the progress is fake.
New thing I started doing: open the COA PDF in a metadata viewer and check Created/Modified timestamps plus author field. If the lab date
I log report ID vs. sample date. If the lab numbers sequentially, a Jan 12 sample with an ID way out of sequence is my first red flag.
My own tracker: 14 COAs from three labs, R² 0.94 between ID order and sample date. The two that broke the trend both had batch mismatches. Not proof, but a cheap filter. Does your checklist include whether the COA’s report ID fits the lab’s public numbering chronology?
I do a dumb but useful check: zoom the vial photo and compare the batch-code font/spacing to another report from the same lab. If one looks pasted-on or from a different printer, that’s my red flag.
I also screenshot the vial + COA side by side into one Notes entry with the date, because paper gets lost between snack drawers here. Sharp question: does your COA list the instrument ID or method number?
If those are missing, I wouldn’t trust a QR or pretty purity number to mean much.
One thing I haven't seen mentioned: ask the lab directly. Most will confirm whether a report ID was issued for a given batch, no vendor involvement needed.
I emailed two labs last month, one replied in a day that the ID didn't match any sample they'd received. That told me more than any PDF metadata. Also worth checking whether the lab is ISO 17025 accredited for that specific test, not just generally.
Login to reply to this thread.
Login / Sign up