What does the Client field on a CoA mean?

It names who commissioned and paid for the test, which frames how you read everything else on the page. On our Janoshik report for the Epitalon batch, the Client is a Telegram handle belonging to the manufacturer, which means the manufacturer sent their own product for testing and received the result.

Does 'third-party tested' mean the test was independent?

It means a third-party laboratory did the analysis. It does not tell you who commissioned it, and those are different things. On our own page the split is: of 58 certificates, 43 were commissioned by the manufacturer, 12 came from the community, and 3 by us. Where a manufacturer commissions the test, the lab is still a third party, but the choice of which vial was sent and what happened to the result sits outside the lab.

What is the void volume comment about?

In chromatography, anything that does not interact with the column washes through immediately, in what is called the void volume. Our Janoshik report states that this unretained peak "was not included in the purity assessment", so the purity figure is a ratio calculated among the retained peaks and not a share of the vial's total mass. The report does not say what the unretained peak contained, so we cannot tell you either. What it is, unambiguously, is a disclosure of how the number was calculated.

Why does purity have three decimal places?

99.154% is a peak-area ratio, and the number of decimals printed does not by itself say anything about the precision of the measurement. Without a stated measurement uncertainty, extra decimals are not extra confidence. Note the contrast with the Liquilabs report, which prints an uncertainty next to about half its analyte results, for example 15.56 mg plus or minus 0.08.

What is the difference between purity and content on a certificate?

They answer different questions. Purity is a proportion: how much of what was measured was the target compound. Content, reported in mg, is a quantity: how much peptide the vial actually holds. A vial can be highly pure and underfilled, or contain the labelled amount of a lower-purity material. A certificate can carry one value without the other.

Is a batch called 'Green' a real batch number?

It is the batch identifier used by that manufacturer, and in this case it refers to a cap colour rather than a pharmaceutical lot number. It works for tracing as long as it is applied consistently and appears identically on the vial, the certificate and the verification record. It is not the same thing as a GMP lot number.

Why does the report not name a method?

Because the requested service was recorded as "Assessment of a peptide vial or vials". The comment about the column's void volume tells you chromatography was involved, but no technique or procedure is named anywhere on the page. Compare that with the Liquilabs report, where the microbiology and elemental rows cite compendial procedures such as USP ` ` or Ph. Eur. 2.6.12, and the peptide rows at least name an in-house one.

What is an acceptable range and why does it matter?

It is the limit a result is judged against, and it turns a number into a verdict. The Liquilabs report states, for instance, endotoxin below 0.001 EU/mg against an acceptable range of 0 to 0.5. Without a range, a result is just a value: you know what was measured, not whether it passed anything.

What does FTIR identification tell me?

It compares the sample's infrared spectrum against a reference and reports a match score, such as 989 out of 1000 in our GLOW report. It is one identity signal. The Liquilabs report pairs it with a retention-time match and, on other reports, a molecular-ion mass from mass spectrometry. It establishes identity by combining signals rather than relying on any single one.

Does a certificate cover the exact vial I received?

No, and this is inherent to how batch certification works. The certificate describes a sample from that batch which was sent to a lab at a specific time. Your vial comes from the same batch, not the same vial. Batch testing rests on the assumption that the batch is homogeneous, which is an assumption rather than a measurement of your vial.

How do I verify a Janoshik certificate?

A Janoshik report carries a task number and a unique key, printed at the bottom, for example 4371ZNKCLG9H. Enter the key at janoshik.com/verify, or open the direct link that combines both, which is what we publish on our lab reports page. The point is that it resolves on the lab's own server, not on ours.

Why does it matter where the certificate is hosted?

Because a PDF hosted by a vendor is a document the vendor controls. A record that resolves on the testing lab's own server lets you compare the certificate against the lab's own record instead of trusting the seller's copy. That is a check against the source, not a guarantee that the sampling or the analysis was sound, but it is the difference between a checkable claim and one you take on faith.

What does a certificate not tell me?

It does not tell you the synthesis route, it does not describe your individual vial, and it is not a safety or clinical assessment. Even a report that tests microbiology, endotoxin and heavy metals against the acceptance ranges printed on it, using compendial procedures, is making a materials-quality statement about a research chemical, not a claim that anything is safe to use.

Why do you publish manufacturer-commissioned certificates at all?

Because they are the certificates that exist, and the alternative is publishing nothing. A manufacturer-commissioned report from a real lab, with a working verification link, still shows that a named lab tested a sample of that product. What it does not show is independent sampling or independent commissioning, so we label it accurately rather than calling all 58 reports 'third-party tested' and leaving it there.

What do the three Liquilabs reports cover?

They are the three we commissioned ourselves, and they cover assay, purity, FTIR and retention-time identity for each peptide in the product, plus microbiology, bacterial endotoxin and seven elemental impurities. The microbiology and elemental rows cite compendial procedures (USP and Ph. Eur.) with acceptance ranges; the peptide rows cite an in-house procedure. They are the only reports on our page where the Client field says Peptidesdirect.io.

Where can I see all of your certificates?

On our lab reports page, organised by product and batch, with the testing lab named and a verification link where the lab provides one. At the time of writing that is 58 reports covering 30 of our products.

How to Read a Certificate of Analysis: A Real Report, Field by Field

Peptide vendors say "third-party tested" and link a PDF. A lot of what a certificate actually reveals sits in the fields that are easy to scroll past.

So rather than describe a hypothetical CoA, this article takes two real ones off our own lab reports page and goes through them line by line. One is a single-page report from Janoshik. Janoshik issued 43 of our 58 published certificates, and all 43 of those were commissioned by the manufacturer. The other is a 24-page report from Liquilabs, one of only three we commissioned ourselves. Side by side, the two show what a certificate does and does not tell you.

TL;DR: the fields that actually matter

Client names who paid for the test. It is the field that reframes everything else on the page, and it is easy to scroll past. On our Janoshik report it is a Telegram handle belonging to the manufacturer, not us. Method is absent here. Our Janoshik report says the test requested was "Assessment of a peptide vial or vials". It never names HPLC or mass spectrometry anywhere. Comments is where the limitations hide. Ours says the unretained peak "was not included in the purity assessment", which narrows what the 99.154% covers. Purity and content answer different questions. 99.154% is a ratio. 10.73 mg is a quantity. A certificate can carry one without the other. The verification key is the only field that lets you check any of it against the lab rather than the vendor.

Report 1: the single-page report

This is Janoshik task #99782, covering an Epitalon batch. You can open it yourself at verify.janoshik.com/tests/99782_4371ZNKCLG9H. Every field it carries:

Field	Value on this report	What it actually tells you
Task Number	#99782	The lab's internal ID. Half of the verification link.
Testing ordered	14 JAN '26	When someone asked for the test.
Sample received	15 JAN '26	When the vial arrived at the lab.
Analysis conducted	19 JAN 2026	When the work happened. Note the gap: order, receipt and analysis are three different dates.
Client	A Telegram handle belonging to the manufacturer	Who paid. See below. This is the field that reframes the whole document.
Sample	Epitalon 10mg	What the submitter said it was. Not a finding.
Manufacturer	Named on the report	Who made it, as stated by the submitter.
Batch	Green	The batch identifier. Here it is a cap colour, not a pharmaceutical lot number.
Sample description	"See picture or pictures attached."	The lab photographs the vial it received. That photo is the sample's identity: a glass vial with a handwritten label.
Tests requested	"Assessment of a peptide vial or vials"	No method is named. Not HPLC, not mass spectrometry.
Results	Epithalon 10.73 mg / Purity 99.154%	Two numbers. One quantity, one ratio.
Comments	"The unretained peak, eluting at the column's void volume, was not included in the purity assessment."	A methodological disclosure that narrows what the purity figure covers. See below.
Signature	Handwritten	A person signed it.
Verification key	4371ZNKCLG9H	The one field that lets you check the rest.

Task Number

Value on this report: #99782
What it actually tells you: The lab's internal ID. Half of the verification link.

Testing ordered

Value on this report: 14 JAN '26
What it actually tells you: When someone asked for the test.

Sample received

Value on this report: 15 JAN '26
What it actually tells you: When the vial arrived at the lab.

Analysis conducted

Value on this report: 19 JAN 2026
What it actually tells you: When the work happened. Note the gap: order, receipt and analysis are three different dates.

Client

Value on this report: A Telegram handle belonging to the manufacturer
What it actually tells you: Who paid. See below. This is the field that reframes the whole document.

Sample

Value on this report: Epitalon 10mg
What it actually tells you: What the submitter said it was. Not a finding.

Manufacturer

Value on this report: Named on the report
What it actually tells you: Who made it, as stated by the submitter.

Batch

Value on this report: Green
What it actually tells you: The batch identifier. Here it is a cap colour, not a pharmaceutical lot number.

Sample description

Value on this report: "See picture or pictures attached."
What it actually tells you: The lab photographs the vial it received. That photo is the sample's identity: a glass vial with a handwritten label.

Tests requested

Value on this report: "Assessment of a peptide vial or vials"
What it actually tells you: No method is named. Not HPLC, not mass spectrometry.

Results

Value on this report: Epithalon 10.73 mg / Purity 99.154%
What it actually tells you: Two numbers. One quantity, one ratio.

Comments

Value on this report: "The unretained peak, eluting at the column's void volume, was not included in the purity assessment."
What it actually tells you: A methodological disclosure that narrows what the purity figure covers. See below.

Signature

Value on this report: Handwritten
What it actually tells you: A person signed it.

Verification key

Value on this report: 4371ZNKCLG9H
What it actually tells you: The one field that lets you check the rest.

The Client field, and what "third-party tested" really means

Start here, because it changes how you read everything else.

On this report, the Client is a Telegram handle, and it belongs to the manufacturer of the peptide. Not to us. Not to an independent auditor. The manufacturer sent their own product to the lab, paid for the test, and received the result. Which vial they chose to send, and what they did with the result, are decisions that sit outside the lab.

That is not fraud, and Janoshik is a real lab doing real analysis. It is what "third-party tested" describes in this case: a third-party lab, commissioned by the first party. The laboratory itself has no stake in the outcome. What manufacturer commissioning does not establish is independent sampling, or independent control over which results get published.

Our own numbers, so you can calibrate

Of the 58 certificates on our lab reports page: 43 were commissioned by the manufacturer, 12 were submitted by the community, and 3 were commissioned by us (the Liquilabs reports below). We do not operate a laboratory, and we have not commissioned testing beyond those three. We do not describe the other 55 as independent.

That breakdown comes from our own report records, and each report carries a Client field you can read for yourself on the linked document.

The Comments field, and why 99.154% is narrower than it looks

The comment on this report reads:

"The unretained peak, eluting at the column's void volume, was not included in the purity assessment."

Unpack that, because it is doing real work.

In chromatography, the sample is pushed through a column that retains compounds according to how strongly they interact with it. Anything that does not interact at all washes straight through in the void volume, arriving in one lump at the very start of the run. Salts, buffer components and other non-retained material end up there.

The comment says that lump was excluded from the purity calculation. The report does not identify what was in it, and that is the honest reading: unretained material commonly includes salts and system-related components, but nothing on the certificate establishes its composition. What the disclosure does tell you precisely is that the 99.154% is a ratio calculated among the retained peaks, not a statement that 99.154% of the vial's mass is Epitalon.

This is the practical reason purity and content are separate values. The purity figure is the target peak's share of the integrated peak area, after the void peak was excluded. The content figure, 10.73 mg, is the quantity claim. On this vial, labelled 10 mg, the measured content came out slightly above the label.

Notice also what the report does not carry: no method, no measurement uncertainty, no acceptance range. Three decimal places do not by themselves establish measurement precision, and this report states no measurement uncertainty.

Report 2: what a fuller report looks like

Now the contrast. This is Liquilabs report LYO-0261, published 29 June 2026, covering a batch of our GLOW blend. The Client field on this one says Peptidesdirect.io, because we ordered and paid for it. It runs to 24 pages where the Janoshik report is one. Its peptide rows all cite the same procedure string, "Peptide Screening 0.1% TFA".

For each of the three peptides in the blend, it reports four separate things:

Test	BPC-157	GHK-Cu	TB-500
Assay (quantity)	15.56 mg [± 0.08]	52.6 mg [± 0.3]	8.68 mg [± 0.04]
Purity	99.5 % [± 0.5]	> 99.8 %	99.0 % [± 0.5]
Identification by spectrum (FTIR)	989 [± 5]	991 [± 5]	987 [± 5]
Identification by retention time	0.996 [± 0.005]	0.989 [± 0.005]	0.996 [± 0.005]

Assay (quantity)

BPC-157: 15.56 mg [± 0.08]
GHK-Cu: 52.6 mg [± 0.3]
TB-500: 8.68 mg [± 0.04]

Purity

BPC-157: 99.5 % [± 0.5]
GHK-Cu: > 99.8 %
TB-500: 99.0 % [± 0.5]

Identification by spectrum (FTIR)

BPC-157: 989 [± 5]
GHK-Cu: 991 [± 5]
TB-500: 987 [± 5]

Identification by retention time

BPC-157: 0.996 [± 0.005]
GHK-Cu: 0.989 [± 0.005]
TB-500: 0.996 [± 0.005]

Then the parts that the single-page Janoshik report does not contain at all:

Section	Test	Method named on the report	Result	Acceptable range
Microbiology	Total aerobic microbial count	USP `<61>` / Eur. Ph. 2.6.12, plate count	0 CFU/g	0 - 1000
Microbiology	Total yeast and mold count	USP `<61>` / Eur. Ph. 2.6.12, plate count	0 CFU/g	0 - 100
Microbiology	Bacterial endotoxin	USP `<85>` / Eur. Ph. 2.6.14, chromogenic	< 0.001 EU/mg	0 - 0.5
Elemental impurities	Arsenic, cadmium, mercury, lead	USP `<232>` / Ph. Eur. 5.20 / 2.4.20	< 0.001 ppm each	0 to 1.5 / 0 to 0.5 / 0 to 1.5 / 0 to 1.5
Elemental impurities	Nickel, vanadium, cobalt	USP `<232>` / Ph. Eur. 5.20 / 2.4.20	< 0.001 ppm each	0 - 25 each

Microbiology

Test: Total aerobic microbial count
Method named on the report: USP <61> / Eur. Ph. 2.6.12, plate count
Result: 0 CFU/g
Acceptable range: 0 - 1000

Microbiology

Test: Total yeast and mold count
Method named on the report: USP <61> / Eur. Ph. 2.6.12, plate count
Result: 0 CFU/g
Acceptable range: 0 - 100

Microbiology

Test: Bacterial endotoxin
Method named on the report: USP <85> / Eur. Ph. 2.6.14, chromogenic
Result: < 0.001 EU/mg
Acceptable range: 0 - 0.5

Elemental impurities

Test: Arsenic, cadmium, mercury, lead
Method named on the report: USP <232> / Ph. Eur. 5.20 / 2.4.20
Result: < 0.001 ppm each
Acceptable range: 0 to 1.5 / 0 to 0.5 / 0 to 1.5 / 0 to 1.5

Elemental impurities

Test: Nickel, vanadium, cobalt
Method named on the report: USP <232> / Ph. Eur. 5.20 / 2.4.20
Result: < 0.001 ppm each
Acceptable range: 0 - 25 each

Three differences worth naming

A named procedure on every line. The microbiology and elemental rows cite compendial procedures (USP and Ph. Eur. chapters). The peptide rows cite an in-house screening procedure, "Peptide Screening 0.1% TFA", which is not a pharmacopoeial method but is at least named. The single-page report names nothing at all.

An uncertainty on about half the rows. "15.56 mg [± 0.08]" carries its measurement uncertainty. Counting the analyte rows on this report, 11 of 22 do. The other 11 carry no interval at all: the GHK-Cu purity reads "> 99.8 %", the three microbiology rows read "0 CFU/g" or "< 0.001 EU/mg", and the seven elemental results read "< 0.001 ppm". "10.73 mg" on the other report is also a measurement; what it lacks is any stated uncertainty at all.

An acceptance range. "0 CFU/g, acceptable 0 to 1000" tells you what the result is being judged against. Without a range, a result is a value with no verdict attached.

What neither report can tell you

Being fair to both documents matters more than scoring points off one of them.

Neither tells you about the vial you received. Both describe a sample that was sent to a lab at a point in time. Your vial is from the same batch, not the same vial. That is the entire basis on which a batch certificate works, and it is an assumption, not a measurement.

Neither tells you the synthesis route. A certificate reports on the finished material. How it was made is not on it, and no purity figure reveals it.

Neither is a safety assessment. Even the 24-page report, which adds microbiology, endotoxin and heavy-metal results against the acceptance ranges printed on it alongside its peptide assay, purity and identity rows, is still only a materials-quality statement about a research chemical. It is not a clinical claim, and none of these products are approved medicines.

A purity percentage is method-dependent. Change the column, the gradient or what you exclude from the integration, and the number moves. This is why the EMA's synthetic-peptide guideline requires impurity methods to be suitable for the threshold they are used against, and addresses co-eluting peaks explicitly. We go through that document in our guide to the guideline.

The five questions worth asking of any certificate

Who is in the Client field?

If it is the manufacturer or the vendor, the lab is still a third party, but the commissioning and the choice of sample are not. If the field is absent, you simply cannot tell who commissioned the test, which is less information than a named manufacturer gives you.

Is a method named?

Distinguish three levels. "HPLC" or "LC-MS" names a technique. "USP <61>" names a compendial procedure. A fully specified, validated method is more than either. "Assessment of a peptide vial or vials" names none of them: it describes a service. All can accompany real work, but only a named procedure lets you ask what it was judged against.

Is there a quantity, not just a ratio?

Purity is a proportion. Content in mg is the amount. A certificate with only a purity figure has not told you how much peptide is in the vial.

What does the Comments field say?

This is where a good lab discloses what it excluded, what it could not resolve, or what it assumed. An empty comments field is not automatically better than a full one.

Can you verify it away from the vendor?

A batch number and a key that resolve on the testing lab's own server let you compare the certificate against the lab's own record, so you are checking with the lab rather than trusting the seller's PDF. That is provenance, not proof that the sampling, the method or the result was right. A PDF hosted only by the seller does not even give you that. Every certificate on our lab reports page carries the lab's own verification link where the lab provides one.

Products and Categories Referenced

The batches discussed above belong to these products. Their certificates, including the two dissected here, are on our lab reports page.

Longevity & Anti-Aginglongevity

Mitochondrial function, NAD+ metabolism, telomere maintenance

Epitalonlongevity

Tetrapeptide (Ala-Glu-Asp-Gly) that activates telomerase, the enzyme responsible for maintaining telomere length. One of the most studied peptides in longevity research, developed by Prof. Khavinson at the St. Petersburg Institute of Bioregulation.

GLOWregeneration

3-in-1 skin peptide blend: GHK-Cu 50mg + BPC-157 10mg + TB-500 10mg. Targets collagen synthesis, tissue regeneration, and skin repair for comprehensive dermatological research.

BPC-157regeneration

Gastric pentadecapeptide (15 amino acids) known for exceptional tissue repair properties. Promotes wound healing, angiogenesis, and cytoprotection across tendons, muscles, gut, and nerves. Over 30 years of preclinical research.

GHK-Culongevity

Naturally occurring copper tripeptide complex for skin regeneration and anti-aging research. Stimulates collagen synthesis, accelerates wound healing, and modulates 4000+ genes. Plasma levels decline with age, making it a key target in longevity research.

TB-500regeneration

Full-length 43-amino-acid Thymosin Beta-4, a naturally occurring repair protein, independently confirmed by a third-party CoA from Janoshik. Promotes cell migration and new blood vessel formation for systemic tissue healing. Especially researched for muscle, tendon, and cardiac repair.

The batches dissected here

Epitalonlongevity

GLOWregeneration

3-in-1 skin peptide blend: GHK-Cu 50mg + BPC-157 10mg + TB-500 10mg. Targets collagen synthesis, tissue regeneration, and skin repair for comprehensive dermatological research.

The peptides inside that blend

BPC-157regeneration

GHK-Culongevity

Frequently Asked Questions

FOR RESEARCH USE ONLY. Not for human consumption. Nothing in this article is medical advice or a therapeutic claim. Certificates of analysis describe material properties of a research chemical and are not safety assessments for any use in humans.

Report 1: the single-page report

The Client field, and what "third-party tested" really means

The Comments field, and why 99.154% is narrower than it looks

Report 2: what a fuller report looks like

What neither report can tell you

The five questions worth asking of any certificate

Products and Categories Referenced

Frequently Asked Questions

What does the Client field on a CoA mean?

Does 'third-party tested' mean the test was independent?

What is the void volume comment about?

Why does purity have three decimal places?

What is the difference between purity and content on a certificate?

Is a batch called 'Green' a real batch number?

Why does the report not name a method?

What is an acceptable range and why does it matter?

What does FTIR identification tell me?

Does a certificate cover the exact vial I received?

How do I verify a Janoshik certificate?

Why does it matter where the certificate is hosted?

What does a certificate not tell me?

Why do you publish manufacturer-commissioned certificates at all?

What do the three Liquilabs reports cover?

Where can I see all of your certificates?

Research context for English-speaking buyers