How to tell if your peptide is actually third-party tested

For a peptide to be honestly described as third-party tested, three conditions have to hold simultaneously. Drop any one of them and the phrase becomes marketing language rather than a verifiable claim.

1. Independent lab. The laboratory that issued the report is a different legal entity from the manufacturer and the seller, with no commercial stake in the result.
2. Batch-specific. The report covers the specific batch the buyer would receive, identified by a batch / lot number that appears on the vial label — not a different batch and not a generic specification.
3. Actually issued. The lab in fact issued a report on this batch (not just appears on the vendor's website as a logo, not just “available on request” without ever materializing).

Every diagnostic below is a different way of checking one of these three.

Look for the “issuing laboratory” line on the COA. It should name a lab, with a real address and a real analyst signature, that is not the seller and not the manufacturer.

Two common patterns fail this test. The first is an in-house QC department whose name is similar to but distinct from the parent company — a sign that the “independent” lab is in fact the same entity. The second is a COA signed only by a title (“Quality Assurance”) with no analyst name and no lab address, which makes the claim of independence unverifiable.

If the issuing lab is named and is genuinely independent, take one extra step: search for the lab itself. A real third-party analytical lab has a website, a physical address, and a list of services it offers. A lab that exists only as a name on a competitor's COA is a flag.

A COA that does not name a batch is not evidence about any specific vial. The minimum signal is: the batch / lot number on the COA matches the batch / lot number printed on the vial label.

A few practical patterns:

• If the vial has no batch number at all, no COA can be tied to it.
• If the COA on the website is older than the vial in hand (manufacture date in the past relative to the vial), it is a different batch.
• If two different products share a batch number, the batch numbering is not unique enough to be informative.
• If the COA “applies to all batches,” it is a specification sheet, not a batch report.

Batch traceability is the cheapest diagnostic to run and it filters out a surprising fraction of vendors who use the phrase “third-party tested.”

A real third-party report covers a defined set of assays. A “third-party tested” claim backed only by a single number (“99% pure”) is weaker than one backed by a structured report. Look for, at minimum:

• Identity by LC-MS, reported as observed mass vs. theoretical mass with a deviation in daltons and ppm.
• Purity by HPLC, reported with the column, mobile phase, gradient, detection wavelength, and the size of the largest impurity in addition to the main-peak area.
• Content by AAA, quantitative HPLC against a reference standard, or UV at 280 nm — reported as a numeric value, not as the labeled dry mass.
• Counter-ion content and water content where applicable.
• Endotoxin by LAL or recombinant Factor C, for any material intended for parenteral use.

A one-page document with only a purity percentage and a signature is technically a COA. It does not establish identity, content, or counter-ion — which means the claim it supports is narrower than “third-party tested” in the way most buyers would read those words. See how to read a peptide COA for a full walk-through.

A signed PDF can be altered after the fact, and the alteration is invisible to the recipient. The strongest version of “third-party tested” is one where the buyer can re-verify the report later, ideally without trusting the issuing lab.

Two patterns satisfy this:

• A public verification URL hosted by the issuing lab, where the original COA can be re-fetched and compared to the PDF the buyer received.
• An on-chain anchor — the COA is content-hashed (SHA-256) and the hash is written to a public blockchain at issuance. Anyone holding the COA can recompute the hash and confirm it matches the on-chain record. Any alteration changes the hash; any alteration is caught. CertikLabs uses this pattern for every report — see how the on-chain anchor works.

A COA that is only a static PDF, with no way to re-fetch or re-verify it after the fact, is weaker by an order of magnitude.

Across the research peptide market, three patterns recur. None of them is necessarily intentional misdirection; all three look similar at a glance and dissolve under the diagnostics above.

1. One COA, many batches. A single, real, independently issued COA from one historical batch is referenced as evidence that all current batches are also third-party tested. Diagnostic 2 catches this.
2. In-house lab renamed. The supplier's internal QC department operates under a separate-sounding name. The address and signatures on the COA tie back to the parent. Diagnostic 1 catches this.
3. Cross-product implication. A COA for product A from an independent lab is published, and the “third-party tested” claim is then made for product B as well, without B ever being tested. Diagnostics 2 and 3 together catch this.

• Is the issuing lab a different entity from the seller, with a real address and analyst name?
• Does the COA name a batch / lot number that matches the vial?
• Does the report cover identity, purity, content — not just purity?
• Is the report verifiable after issuance (public URL or on-chain anchor)?
• Is the manufacture date on the COA consistent with the vial in hand?

Four or five yeses: the claim is real. Two or three: the claim is partial — ask the supplier to fill in the gaps before relying on it. Zero or one: the phrase is being used as marketing, not as evidence.