Once the lot on the vial matches the lot on the PDF, people still misread the assay results. This page is the legend for those numbers. The owner page for “how to walk a COA” is how to read a peptide COA. The portal walkthrough is the batch verification guide. Use the COA checklist at the bench.
Purity is usually a chromatogram, not a grade
On peptide COAs, the headline percentage is often HPLC area-% of the main peak under a stated method. It is a snapshot for that lot, on that instrument setup, on the printed date. Understanding HPLC explains why a second lab can print a different number without either file being fictional.
Treat the COA as the supplier’s reference layer. Your lab still sets acceptance criteria.
Assay is a different question when it appears
Some documents print an assay value (how much target is present by a stated method) next to purity (how clean the chromatogram looks). They can move in different directions. Do not average them. Do not pick the larger one for a quote comparison.
If the PDF only prints one number, do not invent the other in the ELN.
Water, solvents, and “other”
When present:
- Water / moisture describes residual water under the listed method. It is not a storage SOP.
- Residual solvents are only as good as the list the lab actually ran.
- Appearance is a qualitative note. Pair it with what you see, but do not let a “white powder” line override a lot mismatch.
None of these lines convert research material into a medicine.
A working order of operations
- Names and lots match (understanding COAs).
- Method titles are readable (HPLC, MS, KF, etc.).
- Record each numeric field under its own heading in the receiving log.
- Compare to your acceptance SOP — not to a marketplace screenshot.
- File the PDF against the inventory row and the COA hub record when VaultLabs is the supplier. A common worked case is BPC-157 — lot lookup first, calculator only after the lot is accepted.
Specification, result, and limit are three different columns
Most misreadings come from collapsing three columns that a well-built COA keeps apart, and the confusion survives because the words look interchangeable in a summary email.
A specification is the acceptance range the lot was expected to meet. A result is what this lot actually measured. A limit of detection or quantitation is the point below which the method cannot say anything useful. "Not detected" is a statement about the method's reach, not a guarantee of absence — it means nothing was seen above the limit that method can see.
Read them in that order and the document behaves. Read a result as a specification and you will treat a one-off measurement as a standing promise about future lots. Read "not detected" as "none present" and you will over-claim something the instrument never established.
Where a COA prints a result with no corresponding specification, note that gap rather than inferring one. An unbounded result is information, but it is not a pass.
Mistakes that waste qualified lots
- Using purity to justify a SKU substitution.
- Copying a number from lot A onto lot B “because it is the same peptide.”
- Calling a missing residual-solvent panel a pass.
- Reading “research grade” marketing as if it were a method.
For compound-specific receiving, example: retatrutide COA verification.
VaultLabs does not interpret COA numbers for study design.
