Most of your compliance work is a tax. You do it because a platform requires it, it costs you time, and the best outcome is that nothing bad happens. Testing documentation is the exception. A certificate of analysis is the one asset in a research-peptide catalog that pays on both sides of the ledger: it satisfies the reviewer who decides whether your Merchant Center account stays live, and it answers the exact question your buyer is already asking before they add to cart. One artifact, two returns.
That’s why testing documentation deserves more engineering attention than most operators give it. It isn’t a PDF you upload once and forget. It’s a content system, with a URL structure, an indexing strategy, a naming convention, and a set of copy rules that keep it from turning into a liability. Done well, it becomes the most-visited non-product page on your site and the first thing you send to an underwriter. Done badly, it becomes evidence against you.
So here is what’s covered: what a credible certificate contains, how to tell a real one from a decorative one, where to host it so both humans and reviewers find it, and how to reference testing in your copy without stepping into claim territory. If you’re new to this category, start with our peptide compliance guide for the wider framework, then come back here for the documentation layer.
What a Certificate of Analysis Actually Is
A certificate of analysis is a report. It states that a specific quantity of a specific material, drawn from a specific lot on a specific date, was subjected to named analytical methods and produced named results. That is the whole of it. It is a record of measurement, not an endorsement, not a grade, and not a statement about what the material may be used for.
The distinction matters because a great deal of marketing in this category treats a certificate as a seal. It is not a seal. It is a data sheet, and its value comes entirely from the specificity of the data on it.
The Testing Laboratory’s Identity and Independence
A credible certificate names the laboratory that performed the work, gives its physical address, and states its accreditation status. The relevant accreditation for chemical testing is ISO/IEC 17025, the international standard for the competence of testing and calibration laboratories, which covers method validation, equipment calibration, personnel competence, and the content of reported results. Accreditation is granted by a recognized accreditation body against a defined scope, and that scope matters — a laboratory accredited for microbiological testing is not thereby accredited for chromatographic purity work.
If a certificate does not name the laboratory, you do not have a certificate. You have a graphic.
Sample and Lot Identification
The document should identify the material tested with enough precision that the result cannot be silently transferred to a different batch. That means the product name, the lot or batch number, and ideally the quantity received and the condition on receipt. A certificate that says “Peptide X, 99.1%” without a lot number is not traceable to anything. It describes a hypothetical.
Test Date and Receipt Date
Two dates matter: when the laboratory received the sample and when it performed the analysis. Both should appear. A certificate with no date is unusable for the purpose it exists to serve, because the entire point is to tie a measurement to a moment.
Analytical Methods
For synthetic peptides, two methods do most of the work, and they answer different questions.
Reversed-phase HPLC answers “how much of what is in this vial is the intended compound relative to other UV-absorbing species?” Peptide purity is conventionally determined by HPLC with detection at 214 nm, the wavelength at which the peptide bond absorbs, and reported as the area percentage of the main peak relative to total integrated peak area. As Sigma-Aldrich’s technical documentation explains, this measurement separates the target sequence from truncated sequences, deletion sequences, and incompletely deprotected material — but it does not account for water, salts, or counterions, because those do not absorb at that wavelength. Chromatographic method performance is governed by system suitability requirements; USP General Chapter <621> sets the framework for that in a compendial context, including how much a method may be adjusted before revalidation is required.
Mass spectrometry answers a different question: “is this actually the compound it is supposed to be?” HPLC tells you a peak is dominant. It does not tell you the peak is the right molecule. Mass spectrometry confirms identity by measuring molecular mass against the theoretical mass calculated from the sequence. A certificate reporting purity without any identity confirmation is reporting that something is very pure, without establishing what that something is.
The Purity Figure and How It Was Calculated
The number should be accompanied by the method that produced it and the basis of calculation. “99.2% (RP-HPLC, 214 nm, area normalization)” is a statement. “99.2% pure” is a marketing figure. Note also the distinction between purity and net peptide content: purity is a chromatographic ratio, while net peptide content — typically established by amino acid analysis — is the proportion of the total mass that is peptide at all, with the remainder being water, residual solvent, and counterion. The two are multiplicative. A material can be 99% pure by HPLC and still be well under 99% peptide by mass. Certificates that report only the first figure are not wrong, but they are incomplete, and sophisticated buyers know it.
Appearance, Solubility, Water Content
Where relevant, a full certificate reports physical description (typically a white to off-white lyophilized powder), solubility observations, and water content by Karl Fischer titration or loss on drying. These are secondary to purity and identity, but their presence signals that the laboratory ran a panel rather than a single injection.
Signature or Authorization
A named analyst or quality authority, with a title, and a date of authorization. Reporting requirements under ISO/IEC 17025 exist precisely so that a result can be attributed to a competent person operating under a quality system.
How to Read One, and the Tells of a Weak Certificate
Read a certificate in this order: lot number, date, laboratory name, methods, then the number. If the first four are missing or vague, the number is decoration. Specific warning signs, all of which we see regularly in this category:
- No lot number, or a lot number that appears on certificates for multiple different products.
- No date, or a date that has not changed across a year of new inventory.
- A laboratory name with no address, no accreditation reference, and no web presence.
- A purity figure with no method annotation, or a suspiciously round number.
- Chromatogram images that are low-resolution, cropped so axes are unreadable, or visually identical across different products.
- A retention time on the chromatogram that does not match the retention time stated in the text.
- Vendor branding on a document that claims to be third-party work.
- Language about efficacy, benefit, dosing, or human use anywhere on the document. Laboratories report measurements. They do not opine on use.
Third-Party Versus In-House Testing
The two prove different things, and conflating them is the most common documentation error we correct.
In-house testing proves that a process control exists. If you run HPLC on every incoming lot and reject material below a threshold, you have a quality process, and that is genuinely worth something. It demonstrates operational maturity, and it produces data faster and cheaper than sending everything out.
Third-party testing proves the same measurement plus one additional thing: that the party reporting the number has no commercial interest in the number. That independence is the entire premium.
When In-House Is Acceptable
In-house data is reasonable as a supplement, as a process control record, or for high-frequency monitoring where third-party turnaround would be impractical. It is not a substitute for independent verification on the lots you actually sell, and it should never be presented as though it were third party. Labeling in-house work as “third party” is a misrepresentation, and misrepresentation is treated by Google Merchant Center’s misrepresentation policy as an egregious violation that can result in account suspension without prior warning.
Why Payment Underwriting Weighs Independence Heavily
Card network compliance programs — Mastercard’s Business Risk Assessment and Mitigation program and the Visa Integrity Risk Program — place the obligation on acquirers to classify high-risk merchants correctly and monitor what they actually sell. LegitScript’s summary of BRAM and VIRP states that acquirers failing to do so can face fines as high as six figures per transaction, and that for merchants, violations can result in financial penalties and account termination, with a violative merchant included in a database of merchants whose accounts were terminated within the preceding five years — which, as LegitScript puts it, “can make it difficult to acquire a merchant account in the future.”
That penalty structure explains underwriter behavior. An acquirer reviewing a research-peptide merchant is not primarily assessing product quality. They are assessing the probability that this merchant becomes their problem.
Independent documentation reduces that probability estimate in a way self-generated documentation cannot, because the underwriter’s question is not “is this material pure?” but “is this merchant the kind of operator who will misrepresent things?” A named accredited laboratory, with lot-matched reports, answers the second question. In our experience across underwriting reviews, the presence of a coherent third-party testing program is one of the more reliable differentiators between approved and declined applications in this vertical — though we should be clear that this is an observed pattern in submissions we have handled, not a published underwriting rule. For the wider picture on getting and keeping a processor, see our breakdown of payment processing options for peptide merchants.
Lot-Level Versus Product-Level Documentation
This is the section most operators need and least want to hear.
A single undated certificate reused across an entire catalog is worse than having no certificate at all. Not equivalent. Worse.
The reasoning is about what each state communicates to a reviewer. A merchant with no testing documentation presents an absence of evidence. That is a gap. Gaps are remediable, and reviewers see them constantly. A merchant with one certificate applied to forty products presents evidence that does not support what it is being used to support — and once a reviewer notices that the same lot number and the same date appear under every product, the finding is no longer “incomplete documentation.” It is “documentation that misrepresents.”
Those land in different places. The first is a quality gap. The second reads as fabricated evidence, and it recategorizes everything else on the site as suspect. If the merchant misrepresented the testing, the reviewer now has to wonder what else is misrepresented. This is the mechanism by which a well-intentioned shortcut turns a fixable review into a suspension. We cover the recovery path in our guide to Merchant Center suspensions for health products, but avoidance is cheaper than appeal.
The correct model is one certificate per lot, each certificate tied to the lots actually in inventory, with older lots archived rather than deleted. If you cannot test every lot, test what you can and say plainly which lots are covered and which are not. A testing program with honest coverage gaps is credible. A testing program with implied universal coverage and one document behind it is not.
Where to Host COAs So Both Audiences Find Them
Testing documentation that nobody can find does not earn either return. Four rules govern placement.
On the Product Page and on a Per-Lot Page
The product page carries a link to current lot documentation, because that is where the buyer’s question forms. The per-lot page is the canonical record, because lots change and product pages do not. The product page points to the lot page. The lot page persists. This separation is what allows a buyer holding a vial from eight months ago to still verify it.
For how to word the product page around that link without creating a claim, see our guidance on peptide product page copy.
Indexable, Not Gated
Do not put testing documentation behind an email capture, a login, or a support ticket. It is the single highest-intent piece of content on your site, and gating it converts a trust signal into a friction point. It also makes the document invisible to reviewers doing pre-contact research, which is how most platform and underwriting review actually begins. Let it be crawled. Let it rank.
Stable URL Structure With a Searchable Lot Number
The lot number must appear as text in the URL, in the page title, in an on-page heading, and in the body content. That is what makes “[product] lot 24-0917 COA” a query your own page can answer. If the lot number exists only inside an image, it is not searchable by anyone, including you.
File Format and Accessibility
Google’s crawlers extract text from documents; a PDF that is a scanned photograph with no embedded text layer contains no extractable text at all, which makes it unindexable and unsearchable. It is also inaccessible — WCAG 2.2 Success Criterion 1.1.1 requires text alternatives for non-text content, and a scanned certificate is an image of text with no alternative.
The practical answer is to render the certificate’s data as HTML on the lot page — laboratory, lot, date, methods, results in a real table — and offer the signed PDF as a linked artifact alongside it. The HTML serves search and screen readers. The PDF serves the reviewer who wants the signed original. Both audiences get what they need from the same URL.
The Trust Page
The per-lot pages are the data. The trust page is the explanation. It is a single, stable, linked-in-the-footer page that describes how testing works at your company, and it does more underwriting work per word than anything else on your site.
What Belongs On It
- Testing program description. What is tested, at what frequency, against what internal specification, and what happens to material that fails. Name the sampling approach — every lot, every incoming shipment, periodic — and be accurate about it.
- Laboratory partners. Who performs the analysis, their accreditation status, and the scope of that accreditation. If you use more than one laboratory, say which does what.
- Sourcing and handling. Where material originates in general terms, what documentation you require from suppliers, and how incoming material is verified before it enters inventory.
- Storage conditions. Temperature, light, and humidity controls for lyophilized material in your facility, and how orders are packed. This is a materials-handling statement, not a stability claim.
- Quality process. Specification thresholds, what triggers a rejection, how lots are numbered, how records are retained, and for how long.
- Contact route for documentation requests. A named address that receives requests for lot documentation, with a stated response window. Underwriters and platform reviewers use this. So do institutional buyers.
Why One Page Serves Three Audiences
A platform reviewer wants to know whether the site is what it claims to be. An underwriter wants to know whether the operator has controls. A buyer wants to know whether the material is real. All three questions are answered by the same disclosure of process, because process disclosure is the common denominator of credibility. The trust page is also the single URL you paste into a merchant application, an appeal, or a processor questionnaire — which is a practical argument for keeping it at a permanent address and never redesigning it into a slideshow.
Where the trust page sits in the wider account-readiness sequence is laid out in our research peptide compliance checklist.
Referencing Testing Without Making a Claim
This is the hard part, and it is where otherwise careful operators lose accounts.
The governing principle is narrow and useful: purity is a property of the material; efficacy, safety, and suitability for use in a person are not properties of the material. You can describe what a measurement found. You cannot describe what the measurement implies about what the compound does, or for whom.
The FTC’s Health Products Compliance Guidance is explicit that objective claims require substantiation, that implied claims count as claims, and that claims about the level of scientific support — “clinically tested,” “proven” — are themselves objective claims requiring their own substantiation. It also states that disclaimers cannot contradict an express claim. That last point is why “for research use only” at the bottom of a page describing weight loss does not save the page.
FDA has applied the same logic directly to this category, repeatedly. In its December 2024 warning letter to Summit Research Peptides, the agency quoted website product descriptions including “Supports weight management in obesity,” “Promotes better cardiovascular health,” and “Maintain stable blood sugar levels and achieve effective glucose control,” and concluded the products were intended as drugs for human use despite labeling marketing them as “RESEARCH USE ONLY” and “INTENDED AS A RESEARCH CHEMICAL ONLY.”
The position has not softened. In a June 2026 warning letter to Wholesale Peptide, FDA stated: “Despite statements on your product labeling marketing your products for, ‘RESEARCH USE ONLY’ and ‘not for human consumption,’ evidence obtained from your product labeling, including your website establishes that your products are intended to be drugs for human use.” The letter cited sections 301(d) and 505(a) of the FD&C Act. Notably, some of the claims FDA quoted were framed in research language — “Research shows Prostamax can improve bladder control” — which did not protect them. Attributing an outcome claim to research does not stop it from being an outcome claim.
Permitted Phrasings
- “Purity: 99.1% by RP-HPLC (214 nm). Lot 24-0917. Certificate available.”
- “Identity confirmed by mass spectrometry against theoretical molecular weight.”
- “Each lot is analyzed by an independent ISO/IEC 17025-accredited laboratory before release.”
- “Certificates of analysis are published per lot and are publicly accessible.”
- “Supplied as a lyophilized powder for laboratory research use only. Not for human or veterinary use.”
- “Full analytical documentation, including chromatograms, is available on the lot page.”
Prohibited Phrasings
- “Pharmaceutical grade” or “medical grade” — undefined terms that imply an approval status that does not exist.
- “Clinically tested” or “clinically proven” for an analytical purity test. Chromatography is not a clinical trial.
- “Lab tested for safety” — the test measured composition, not safety.
- “Tested pure, so you can dose with confidence.” Any dosing reference converts the page.
- “Safe because it is third-party tested.”
- “Verified effective,” “research-backed results,” or any construction tying testing to an outcome.
- “Research shows [compound] improves [outcome]” — research framing does not neutralize an outcome claim.
- “Human grade,” “injection ready,” “sterile and ready to use.”
The same discipline applies in email, where sequencing and repetition make drift easy; our guide to compliant peptide email flows covers how to reference documentation in lifecycle messaging. For which surfaces enforce which rules, see the current peptide policy map.
Displaying COAs Without Creating a New Problem
A certificate is safe in isolation. It becomes unsafe by adjacency.
Reviewers and regulators read pages as wholes, not as isolated elements. The FTC guidance is direct that context — product names, images, surrounding copy — can convey a claim that no sentence states. So a testing badge placed next to outcome language does not neutralize the outcome language. It strengthens it, by supplying apparent substantiation. “Third-party tested” sitting beside “supports fat loss” reads as a single compound claim: tested, therefore it works. You have made the violation more serious, not less.
Practical rules for display:
- Keep testing badges and certificate links away from any benefit, outcome, condition, or dosing language. If such language exists on the page, remove it; do not try to balance it.
- Do not place testimonials, before-and-after imagery, or user photographs on or near a lot page. That is the exact evidence pattern FDA cited in both letters above.
- Do not caption a certificate with anything a laboratory would not write. “Verified quality you can trust” is your voice, not the laboratory’s, and it converts a data sheet into a promise.
- Never imply that testing establishes safety for human use. Testing establishes composition. Safety for human use is established by clinical evaluation and regulatory approval, neither of which applies to research-use-only materials.
- Keep research-use-only framing on the lot page itself, not only on the product page. The lot page is independently indexable and will be landed on directly.
The underlying risk is a specific inference: tested means checked, checked means safe, safe means fine to use. Every element of your presentation should break that chain rather than complete it.
A Structure You Can Copy
Here is a concrete architecture. Adapt the naming, keep the shape.
| URL | Purpose | Primary audience |
|---|---|---|
| /quality/ | Trust page: program, labs, sourcing, storage, process, contact | Reviewers, underwriters, buyers |
| /quality/testing-methods/ | Plain-language explanation of HPLC and MS, and how to read a certificate | Buyers, search |
| /coa/ | Index of all lots, filterable by product, sortable by date | All |
| /coa/[product-slug]/ | All lots for one product, newest first, archive retained | Buyers, search |
| /coa/[product-slug]/lot-[number]/ | Canonical lot record: HTML data table, chromatogram image with alt text, linked signed PDF | All |
| /quality/document-requests/ | How to request documentation not published, with response window | Underwriters, institutional buyers |
Supporting conventions:
- Lot page title: “[Product] — Lot [number] — Certificate of Analysis.” Lot number in the H2 and in the first paragraph.
- Data as HTML. Laboratory, accreditation, receipt date, test date, method, wavelength, purity, identity confirmation, water content, appearance — each a row in a real table.
- PDF as artifact. Linked, text-layered, named predictably: product-lot-number-coa.pdf.
- Chromatogram images at readable resolution, with alt text describing what is shown.
- Never delete a lot page. Mark superseded lots as archived and keep the URL live. Removing history is the single fastest way to look like you are hiding something.
- Product pages link down; lot pages link up to the trust page. No orphans.
- Research-use-only statement present on every lot page and every product page.
Build it once and it maintains itself at roughly one new page per lot, which is a data-entry task rather than a marketing project.
Frequently Asked Questions
What is a peptide COA?
It’s a laboratory report stating the analytical results for one specific lot of material. A complete one tells you who tested it and what that laboratory is accredited for, the product and lot number, the receipt and test dates, the analytical methods used, and the results, typically purity by reversed-phase HPLC and identity confirmation by mass spectrometry, sometimes with appearance, solubility, and water content alongside. Hold on to what that actually is: a record of measurement on one batch. It isn’t a certification, it isn’t a grade, and it is not any statement about permitted use.
How do I verify third party testing for peptides is genuine?
Start with the laboratory. The certificate should name a real one, with a physical address and verifiable accreditation, and that accreditation scope needs to cover chemical or chromatographic testing. Then confirm the lot number on the certificate matches the vial label. Check that the date is recent, and that different products carry different lot numbers and different dates. Look at the chromatogram itself: the axes should be readable, and retention times should match the text. Vendor branding on a supposedly independent report is a strong warning sign. And when you aren’t sure, contact the laboratory directly.
Does a peptide purity certificate mean the material is safe?
No, and if you take one thing from this article, take this one. A purity certificate reports composition: what proportion of the sample is the intended compound, and whether the molecular identity matches. It says nothing about safety, and nothing about suitability for use in a person. Research-use-only materials have not been evaluated or approved for human use, and no analytical result changes that. So if your copy presents testing as evidence of safety, you’ve created a claim that neither the laboratory nor the data supports.
Is HPLC purity the same as peptide content?
No. HPLC purity is measured at 214 nm, where the peptide bond absorbs, and it’s the area percentage of the target peak relative to total UV-absorbing material. It excludes water, salts, and counterions, because those don’t absorb at that wavelength. Net peptide content, usually determined by amino acid analysis, is the proportion of total mass that is peptide. The two figures multiply, which is the part that gets missed. Material can be 99% pure by HPLC while being substantially less than 99% peptide by weight, and complete certificates report both.
Can I use one COA for my whole product line?
No, and it will hurt you more than publishing nothing at all. A single certificate reused across a catalog is obvious to any reviewer who opens two of your product pages, and it reframes your documentation from incomplete to misrepresented. Google’s misrepresentation policy treats such violations as grounds for suspension without prior warning. So publish one certificate per lot, retain the archived lots, and if your program doesn’t yet cover everything, say plainly which lots are covered. Honest partial coverage is credible. Implied universal coverage is not.
Should COAs be public or available on request?
Public. Gating testing documentation behind a form or a login removes it from the audiences it exists to serve. Buyers searching for purity evidence never see it. Search engines can’t index it. And platform reviewers or underwriters doing pre-contact research simply conclude it does not exist. Publishing it as indexable HTML with a linked signed PDF costs you nothing and turns your highest-intent content into a discoverable asset. Keep a documentation request route as well, for material you haven’t published, but make published documentation the default rather than the exception.
How should I mention testing in ad copy?
Describe the measurement and stop there. “Analyzed by an independent accredited laboratory” and “purity reported by RP-HPLC, certificate published per lot” are factual statements about process and composition, which is exactly what you want. What to avoid: “pharmaceutical grade,” “clinically tested,” “lab tested for safety,” or any construction that connects testing to an outcome, a benefit, or use in a person. And one rule to hold above the rest, never place a testing reference adjacent to benefit language. The combination reads as a substantiated efficacy claim and is treated more seriously than the benefit language alone.
Where This Leaves You
Testing documentation is the rare compliance asset that does not need a business case beyond itself. The buyer who searches for a lot number is the buyer closest to purchase. The reviewer who checks your quality page is the reviewer deciding whether your account continues. Building one system that answers both is not an efficiency trick — it is the natural shape of the problem, because credibility with a regulator and credibility with a customer are the same underlying property viewed from two directions.
The work is unglamorous: a URL convention, a table per lot, an honest description of what you test and what you do not, and copy discipline that holds the line between composition and outcome. None of it is difficult. All of it compounds.
This article is general information about advertising and platform compliance practice and is not legal advice; consult qualified counsel for your specific situation.
If you want the documentation architecture, trust page and product copy built as one system that survives Merchant Center review and processor underwriting, iComChain works with research-peptide and supplement brands as a specialist peptide marketing agency handling exactly this problem. Book a free 15-minute consultation and we will review your current testing documentation and tell you where it stands.
Sources
Every policy and regulatory claim in this article is drawn from the primary documents below, checked on 15 August 2026. Platform policies change without notice; verify against the current version before acting.
- ISO/IEC 17025 — ANAB / ISO
- Sigma-Aldrich’s technical documentation — Sigma-Aldrich
- USP General Chapter <621> — USP
- Google Merchant Center’s misrepresentation policy — Google
- LegitScript’s summary of BRAM and VIRP — LegitScript
- WCAG 2.2 Success Criterion 1.1.1 — W3C / WCAG
- Health Products Compliance Guidance — U.S. Federal Trade Commission
- Summit Research Peptides — U.S. Food and Drug Administration
- June 2026 warning letter to Wholesale Peptide — U.S. Food and Drug Administration