Most research compound suppliers claim third-party testing. Far fewer publish the certificates, and fewer still explain how to read one. A certificate you cannot interpret is not evidence — it is decoration.
This is a walkthrough of a real Janoshik Analytical certificate, using one of ours. Every certificate we publish follows the same layout, so once you can read one you can read all of them.
The header block
Task Number. The laboratory's internal reference for the analysis, for example #204680. This is the number Janoshik uses to identify the job, and it appears in the URL of the published report.
Testing ordered / Sample received. Two separate dates, and the gap between them matters. Testing ordered is when the analysis was commissioned; sample received is when the physical material arrived at the laboratory. A certificate showing a sample received months before the batch was manufactured is incoherent, and that incoherence is one of the easier ways to spot a fabricated document.
Client. Who commissioned the test. This is frequently the manufacturer or a trading company rather than the retailer you are buying from — a reseller usually inherits certificates from further up the chain. It is worth knowing that, and worth being suspicious of any retailer who obscures it.
Manufacturer. Who produced the material. On our certificates this names the manufacturer of the batch directly.
Batch. The manufacturing lot, for example 20260712. This is the single most important field for you, because it is the one you can check against the vial you were sent. A certificate for a different batch tells you nothing about the material in your hand.
Tests requested
This states what the laboratory was asked to do. Ours read "Common GLP-1 peptide blind test (Semaglutide, Tirzepatide and Retatrutide)".
The word blind is doing real work there. In a blind test the laboratory is not told which of the candidate compounds the sample is supposed to be — it identifies the material itself from a panel. A non-blind test asks the narrower question "is this retatrutide, and how pure", which cannot detect a sample that is actually something else entirely. Blind testing is the stronger claim.
Results
Two figures, and they answer different questions.
The quantity — for instance Retatrutide 55.01 mg — is the mass of the identified compound the laboratory measured in the vial. This is a measurement, not a restatement of the label, and it will rarely match the nominal figure exactly.
Fill varies in both directions, and that is worth understanding properly. Lyophilised material is hygroscopic, a little is unavoidably retained on the stopper and vial walls, and filling equipment has tolerances that are proportionally larger at small masses — a 10mg vial is harder to fill precisely than a 60mg one.
Across our own current batch the measured quantities range from roughly 5% below the nominal figure to roughly 17% above it, and the smallest vials show the largest proportional variance. Every one of those figures is published on our Certificates of Analysis page, including the ones that came in under.
What matters is not that every vial lands exactly on its nominal figure — no filling process does that — but that the actual quantity is measured, published, and checkable against the vial you received. A deviation you can see and interrogate is a categorically different thing from a claim nobody has tested. What would warrant questions is a large shortfall, or a consistent one-directional bias across a whole range.
The purity — for instance 99.640% — is the proportion of the material that is the identified compound, with the balance being related substances, residual synthesis reagents and moisture. This is what a claim like "99% purity" refers to, and a certificate is the only thing that substantiates it.
Note that not every certificate reports both figures. Some analyses are commissioned for quantification only and state a mass without a purity percentage. If a purity figure is absent from a certificate, no purity claim should be made on the strength of it.
Analysis conducted
The date the work was actually carried out. Combined with the batch number, this tells you how recently the material in front of you was characterised.
The verification key
At the foot of every certificate is a unique key, for example FCHMILTLKGHR, alongside the instruction to verify at janoshik.com/verify.
This is the part that matters most, and the part most people skip. Enter the key on the laboratory's own website and it returns the report as the laboratory holds it. If the document you were shown has been altered in any way — a purity figure nudged upward, a compound name changed, a batch number swapped — the version on the laboratory's domain will not match.
A PDF or screenshot proves nothing on its own. Both are trivially editable. The verification key, checked on the laboratory's own site, is the only step that turns a certificate into evidence.
What a certificate does not tell you
Worth being clear about the limits. A certificate characterises the sample that was submitted, from one batch, on one date. It says nothing about material from a different batch, and nothing about how the material has been stored or handled since. That is why batch-level certificates matter and why a single certificate reused indefinitely across a whole catalogue should be treated with suspicion.
It also does not speak to anything beyond identity, quantity and purity. Those are the questions analytical testing answers, and they are the only questions we make claims about.
Our certificates are published in full at Certificates of Analysis, with the measured quantity, purity figure and verification key for every one.
All compounds are supplied strictly for laboratory research and analytical purposes. Not intended for human or veterinary use.