What Is a COA? Certificates of Analysis Explained

What is a COA? A COA, or certificate of analysis, is a document issued by a testing laboratory or a manufacturer’s quality-control unit that reports the measured results of named tests on one specific sample or lot of a material, next to the specification each result is judged against. A typical COA has seven parts: the issuer, the sample and lot, the dates, the test methods, the specifications, the results, and a signature or verification code. It describes the sample that was tested, not the material in general.

This explainer covers what a COA is for, who issues one, what is printed on it, what it cannot tell a reader, and how it differs from a specification sheet, a certificate of conformance and a safety data sheet (SDS). For a field-by-field walkthrough of two real peptide reports, see how to read a peptide COA.

COA and related documents at a glance

Source: general quality-documentation practice; the SDS format follows the Globally Harmonized System (GHS) as adopted in the US OSHA Hazard Communication Standard (29 CFR 1910.1200).

Research materials: Vortex Research publishes its third-party lab reports, each with a link to the lab’s own verification portal, at /coa/.

What a COA is for

A certificate of analysis exists so that someone who did not run the tests can see what was measured, how, and with what result. Chemical suppliers, reference-material producers, food and ingredient manufacturers and pharmaceutical manufacturers all issue them. In each case the COA links one physical lot to a set of numbers.

For a laboratory, the COA does three jobs:

• Identity. It records evidence that the material is what the label says it is.
• Quality. It records how pure the material is and, where tested, what contaminants were looked for and whether they were found.
• Traceability. It records which lot was tested, when, and by whom, so that a result in a notebook can be traced back to a specific batch of reagent.

The third job is easy to overlook. If an experiment gives an unexpected result months later, the lot number on the COA is how a researcher finds out whether the reagent changed. That is why a COA with no lot reference is of limited use, however good the numbers look.

Who issues a certificate of analysis

COAs come from two kinds of issuer, and the difference matters.

The manufacturer’s own quality-control unit

Most COAs in commerce are in-house documents. The manufacturer tests its own lot against its own specification and signs the result. This is normal practice and is the basis of most chemical catalogues. Its limit is independence: the same organisation makes the material, runs the tests and reports them.

An independent third-party laboratory

A third-party COA comes from a lab with no stake in the result. The lab receives a sample, runs the requested tests and issues a report in its own name. Many such labs hold ISO/IEC 17025 accreditation, the general international standard for the competence of testing and calibration laboratories. Accreditation covers a defined scope of methods, so it is worth checking that the tests on a given report fall inside that scope.

A third-party report also names its requester: the company that sent the sample. In the research-peptide market, the requester is sometimes the seller and sometimes the wholesale facility that made the batch. Both kinds are real lab data. The second kind is often called a partner-facility report, and its link to the vial on a shelf rests on the batch code and the seller’s statement rather than on the seller having sent the sample. A seller who publishes both kinds should label which is which.

What is on a COA: the seven parts

Source: Vortex Research Editorial Team, based on the structure of published third-party peptide reports and on pharmacopoeial practice, where identification, purity and assay are separate tests with separate acceptance criteria [1].

The specification column deserves a closer look. A result only means something next to a limit. “Purity 99.94%” says little on its own; “purity 99.94% against a specification of at least 95.0%” says the lot met its limit with room to spare. Some fields have no limit and are marked “report only”. Net peptide content is often reported that way.

Pharmacopoeial specifications themselves are not uniform. A review of compendial peptide monographs found large differences between the European Pharmacopoeia and the United States Pharmacopeia in how identification, purity and assay are specified, and some inconsistency within a single pharmacopoeia [1]. Two COAs for the same compound can therefore apply different limits and still both be correct.

What a COA for a research peptide usually includes

For synthetic peptides, the core tests are:

• Identity, ideally by mass spectrometry. Electrospray ionisation produces intact, often multiply charged ions of peptides and proteins, so the measured mass can be compared with the mass calculated from the sequence [2]. Some reports instead compare HPLC retention time with a reference standard, which is a weaker test.
• Purity by reversed-phase HPLC with UV detection, reported as area percent. It shows how much of the peptide-related signal comes from one component. Our explainer on HPLC peptide purity covers how that figure is calculated.
• Net content, the amount of peptide actually present, which purity does not measure. A powder also holds counter-ions and bound water.
• Contaminant tests where performed: bacterial endotoxin (USP <85>), sterility (culture-based USP <71> or a rapid PCR screen) and elemental impurities by ICP-MS (USP <233>). Our endotoxin testing explainer covers the first of these.

Each test answers a different question. The impurities that purity testing is designed to catch in synthetic peptides include deletion and insertion sequences, diastereomers from racemisation, side-chain protection adducts, oxidised forms and dimers. Contamination by unrelated peptides has also been documented [3].

What a COA cannot tell you

A COA is precise about a narrow thing. Four limits apply to any report:

• It describes the sample, not the supply. A lab tests what it receives. Third-party reports usually say so. The ILS Laboratories reports published by Vortex state that sample names, lot numbers and labelled content are provided by the submitting party and have not been independently verified, and that the certificate relates only to the sample tested.
• It is a snapshot. A COA shows the state of a lot on the test date. It says nothing about storage since then. In a comparison of 30 liquid mycotoxin reference standards from 10 manufacturers, 30% of suppliers had standards outside the acceptance limits for their stated concentration, and one product showed only 80% of its target value; the authors called for continuous long-term stability monitoring and for HPLC and LC-MS data to be shown on the certificate of each released batch [4].
• It only covers the tests listed. A report without an endotoxin result says nothing about endotoxin. A reader should not assume tests that are not printed.
• It depends on the method. A purity figure belongs to one column, one gradient and one wavelength. Different methods can resolve different impurities.

COA vs specification sheet vs SDS

These three documents are often filed together and often confused.

Specification sheet

A specification sheet sets out what a product is supposed to meet: identity criteria, a minimum purity, appearance, a content range, and the methods used to check them. It is a promise about the product. A COA is the evidence for one lot. A good COA repeats the specification next to each result so the two can be compared on one page.

Certificate of conformance

A certificate of conformance (CoC) is a signed statement that a lot meets its specification. It usually gives no measured values. A CoC asks the reader to trust the issuer’s judgment. A COA shows the numbers so the reader can judge.

Safety data sheet (SDS)

An SDS is a hazard-communication document, not a quality document. Under the Globally Harmonized System, which US law adopts in the OSHA Hazard Communication Standard, an SDS has 16 sections in a set order, covering identification, hazards, composition, handling and storage, exposure controls, physical and chemical properties, stability and reactivity, and disposal, among others. It applies to a substance in general, not to one lot, and it reports no purity or identity results. A laboratory usually needs both: the SDS for handling the material and the COA for knowing what the material is.

How reference standards fit in

Every measured value on a COA is relative to something. HPLC identity compares a sample with a reference standard. Net content is calculated against a standard of known content. The quality of the reference standard therefore sets a ceiling on the quality of the COA.

Establishing a peptide reference standard is its own analytical project. Work from the United States Pharmacopeial Convention describes a two-step value assignment: a mass-balance approach first assigns a quantitative value to bulk peptide material, and that material then serves as the standard for assigning a value to the vialed lots. Identity is confirmed by NMR, mass spectrometry and chromatography, and content and impurities are assessed by HPLC and gas chromatography. Results from several labs are combined [5]. A routine COA does not repeat that work, but it depends on it.

A real COA, part by part

The table maps the seven parts onto one third-party report published on the Vortex COA page: a MOTS-C 10 mg report commissioned in Vortex’s name.

Source: the ILS Laboratories report as published on vortexresearch.net/coa/, read October 21, 2026.

Two details on this report show the limits discussed above. The lot field names a cap colour rather than a batch code, which is a weaker link between report and batch. And identity was confirmed by HPLC retention time, not by a measured mass. Both are visible only because the full report is published.

How to check that a COA is genuine

A PDF or image on a seller’s website is a copy. The original lives with the issuing lab. A quick check:

1. Find the report number, access code or QR code.
2. Go to the lab’s own website by typing its domain, rather than only following a seller’s link.
3. Retrieve the lab-hosted original and compare every field: product, lot, dates, methods, results and chromatogram.
4. Check that the lab exists independently, with its own address and contact details.

Any difference between the seller’s copy and the lab’s copy is disqualifying. Public control laboratories that analyse seized peptide preparations rely on LC-MS/MS for identification and on validated UHPLC methods for quantification [6], and a market-surveillance study of falsified peptide products bought from suspected illegal internet pharmacies found wide variation in peptide amount per unit and low purity in some samples [7]. Documents that cannot be verified offer no protection against either problem. For supplier-level checks beyond the COA, see how to evaluate a research peptide supplier, and for what a purity figure does and does not mean, see what “high purity” means for research peptides.

Frequently asked questions

What does COA stand for?

COA stands for certificate of analysis. It is a document that reports the measured results of named tests on one sample or lot of a material, alongside the specification each result is judged against. It is issued by a testing laboratory or by the manufacturer’s quality-control unit.

Who issues a certificate of analysis?

Either the manufacturer’s own quality-control unit or an independent third-party laboratory. A third-party report names the lab and the requester, the company that sent the sample. Many testing labs hold ISO/IEC 17025 accreditation for a defined scope of methods.

What is the difference between a COA and an SDS?

A COA is a quality document for one lot: it reports identity, purity and other measured results. An SDS is a hazard-communication document for a substance: it gives handling, storage, exposure and disposal information in 16 set sections. A laboratory usually needs both.

Does one COA cover all of a supplier’s stock?

No. A COA applies to the sample tested, and through its lot code to the batch that sample came from. It says nothing about other batches or about storage after the test date. Matching the lot code on the vial with the lot code on the report is the link a reader should check.

What does an example certificate of analysis include?

A complete peptide COA names the product and lot, the testing laboratory, the test date and a report number. It then lists each test with its method, its specification and the measured result. Typical peptide tests are HPLC purity, identity by mass spectrometry and net content, and some reports add endotoxin, sterility or heavy metals. Real examples are summarised on the Vortex lab report data page.

How do you get a certificate of analysis from a peptide supplier?

Ask for the COA that matches the lot number on the vial, not a generic or sample report. A useful COA is lot-specific, comes from a named laboratory and can be checked against the laboratory’s own records, for example through a report number or a verification portal. Vortex publishes its reports at /coa/.

Research materials: Vortex Research supplies peptides as sealed lyophilised research reagents. Browse the catalogue at /shop/. Published lab reports, with lab-portal verification links, are at /coa/.

References

1. Vergote V, Burvenich C, Van de Wiele C, De Spiegeleer B. Quality specifications for peptide drugs: a regulatory-pharmaceutical approach. J Pept Sci. 2009;15(11):697-710. PMID 19750489. doi:10.1002/psc.1167
2. Fenn JB, Mann M, Meng CK, Wong SF, Whitehouse CM. Electrospray ionization for mass spectrometry of large biomolecules. Science. 1989;246(4926):64-71. PMID 2675315. doi:10.1126/science.2675315
3. D’Hondt M, Bracke N, Taevernier L, et al. Related impurities in peptide medicines. J Pharm Biomed Anal. 2014;101:2-30. PMID 25044089. doi:10.1016/j.jpba.2014.06.012
4. Steiner D, Bartók T, Sulyok M, et al. Global perspectives on mycotoxin reference materials (Part I): insights from multi-supplier comparison study including aflatoxin B1, deoxynivalenol and zearalenone. Toxins (Basel). 2024;16(9):397. PMID 39330855. doi:10.3390/toxins16090397
5. McCarthy D, Han Y, Carrick K, et al. Reference standards to support quality of synthetic peptide therapeutics. Pharm Res. 2023;40(6):1317-1328. PMID 36949371. doi:10.1007/s11095-023-03493-1
6. Vanhee C, Janvier S, Desmedt B, et al. Analysis of illegal peptide biopharmaceuticals frequently encountered by controlling agencies. Talanta. 2015;142:1-10. PMID 26003685. doi:10.1016/j.talanta.2015.04.022
7. Janvier S, Cheyns K, Canfyn M, et al. Impurity profiling of the most frequently encountered falsified polypeptide drugs on the Belgian market. Talanta. 2018;188:795-807. PMID 30029448. doi:10.1016/j.talanta.2018.06.023

Methods note: written by the Vortex Research Editorial Team. Report data were read from the lab-issued document published on vortexresearch.net/coa/ on October 21, 2026. Literature was selected from PubMed for analytical and quality-control topics, and every PMID was checked against its PubMed record and abstract. USP general chapters, ISO/IEC 17025 and 29 CFR 1910.1200 are cited by name.

Vortex Research products are sold for in-vitro laboratory research by qualified researchers. Not for human or veterinary use. Not a drug, dietary supplement or cosmetic. Statements have not been evaluated by the FDA.