Reading a COA: Spotting Quality vs Bunk

A certificate of analysis is supposed to answer one narrow question: does this vial actually contain what the label says, at the purity the label claims? That's the whole job. It isn't a marketing document, a design flourish, or a box a compliant vendor checks once, it's a specific lab's record of what a specific sample tested as, on a specific instrument, on a specific day. The problem is that a PDF with a lab's logo and a wall of numbers is easy to produce and just as easy to reuse, copy, or misattribute, and most buyers have never been shown what actually separates a real one from paperwork that only looks credible.

What HPLC actually measures, and where it stops

High-performance liquid chromatography (HPLC) works by pushing a sample through a column that separates its components based on how they interact with it, then measuring how much material comes off at each point. That's exactly why it's the default method logged for a standard peptide test: Janoshik Analytical, a third-party lab that tests peptides, SARMs, and related compounds for purity and identity, lists "Testing method: HPLC" directly on its own BPC-157 analysis order page. Run against a known reference standard, HPLC answers two questions with real precision, how much of the labeled peptide is actually in the vial, and what percentage of everything the instrument detected is that peptide versus something else. What it doesn't answer on its own is what that "something else" actually is.

That's the job mass spectrometry does instead. Janoshik sells LC-MS (liquid chromatography-mass spectrometry) as a separate add-on specifically because it's a different kind of test: as the lab's own service page puts it, "LCMS testing can detect contamination of peptide products with other peptides and degraded peptide," identifying an unknown peak by its mass rather than just measuring its size. The same logic shows up again in the lab's CHNS elemental test, described as "the golden standard in net peptide content (NPC) determination" by directly measuring a sample's nitrogen and sulfur content rather than inferring purity from a chromatogram. None of these methods substitute for each other. A report built on one of them is answering a narrower question than a report built on two or three.

What a legitimate COA has to show

• A measured purity percentage and a measured amount per vial or dose, not a round number presented as a marketing claim with no analytical run behind it.
• The identity-confirmation method actually used, since HPLC alone confirms how much of a known reference compound is present, while mass spectrometry (or an added contamination screen) is what's required to identify an unlabeled peak.
• A sample or batch reference tying the document to the specific product in hand, not a template a vendor could reuse across every batch it has ever sold.
• The date the sample was actually tested, not just the date the document was formatted or printed.
• The name of the testing lab itself, printed on the certificate, not just "third-party tested" asserted in the vendor's own copy.
• Some way to independently check the report against the lab's own records. Janoshik, for one, publishes a subset of its results publicly and runs a dedicated verification lookup by task number and a unique key, so a buyer isn't just trusting a PDF that could have been edited after the fact.

What actually shows up on a bunk certificate

None of the failure modes below require a chemistry background to catch once you know to look for them. The most common is a document with no sample or batch reference anywhere on it, an image or PDF generic enough to describe any bag of powder a vendor has ever shipped, not the specific vial now in front of you. A close second is a report with no testing lab named on it at all, purity numbers with nothing behind them a buyer could actually check. The one that takes the closest look: a COA whose compound name, CAS number, or sequence doesn't match the product it's attached to, sometimes because it was lifted from a different batch's report and the header was never updated. None of that means every thin-looking certificate is fake. It means a certificate missing any of those fields hasn't actually verified the vial in front of you, whatever the listing next to it claims.