How to Read a Peptide COA

You've sent the cash, the courier has dropped off the little vial of white powder, and now you're staring at it, wondering if your money went into a nice round hole. You can't read labels, you can't taste molecules. What is in this vial? A top shelf research peptide, or a fancy lie?

The Certificate of Analysis is the one document that can help you answer that question. This guide breaks down what you need to know to understand exactly what's in that vial in simple terms. Every COA discussed has actual figures from real reports used to highlight the details that matter.

What a COA actually is

A Certificate of Analysis is a lab report for a particular batch of a particular compound. Much like the nutrition facts on the back of a food package, but with chemistry instruments doing the measuring instead of a dietician.

A COA serves three distinct roles, and most people only ask about one of them.

What is it. Is the compound in this bottle actually the peptide you were promised, or something else entirely?

How pure is it. Of all the materials in this vial, how much of it is the peptide you're paying for?

How much is there. Does a vial labeled to contain ten milligrams of peptide actually have ten milligrams of peptide in it?

Identity, purity, and content. Three questions, and three different pieces of information most COAs list on the page. A seller who correctly addresses one, and hides details on the other two is telling you only part of the story.

The two machines behind every real COA

Every real COA has been tested on two machines, usually in two different ways. Knowing what they do and what results to look for makes most COAs much easier to read.

HPLC testing identifies purity. High Pressure Liquid Chromatography separates a compound for testing like an airport security line. Everything in the sample is run through a narrow column at high pressure so individual compounds are separated by the time it takes to move through the column. A detector at the end of the column counts the amount of each substance, generating a graph of peaks representing the contents of the sample. Purity is then determined by the size of the largest peak relative to the size of all the other peaks in the graph. If your peptide peaks at 99 percent of the total amount registered, it's a 99 percent pure sample.

Mass spectral analysis confirms identity. While having a pure sample is nice, purity is only half the battle. At best, an HPLC test on a sample that's 99 percent incorrect peptide will tell you that it's 99 percent incorrect. Mass spectroscopy is the test that confirms the sample is the correct peptide. Every material has a measurable weight based on the atoms it's made of, and a mass spec will cross reference the weight of the molecules in a sample against a set of known weights to identify what it is.

Both tests are essential for any reliable COA. A report without one of these is like a restaurant that only advertises its vegetarian dishes but never actually lists what vegetables are available.

Reading the HPLC section

Purity percentages. The headline statistic for any peptide COA, and it has a minimum standard for research materials. Most sellers offering research peptides should be able to post figures of 98 percent or higher for purity, and the best companies frequently post numbers higher than 99 percent. Anything lower than 95 percent and you're in dangerous territory, and you should take another long hard look at any COA that lists 100 percent purity. In reality, chemistry rarely achieves absolute perfection, and a suspiciously round number often indicates that the second decimal place has been swept under the rug for marketing purposes.

The chromatogram. A visual representation of the peaks discussed earlier. Ideally, this graph should show one large peak for the peptide of interest, and a smaller, flat line for everything else in the sample. Several large peaks will indicate a mixture of materials, rather than one dominating peptide.

Retention time. A measure of how quickly a substance moves through the chromatography column in minutes, and another reality check for the COA. Like a sprinter running the same race at the same track, the retention time for an authentic sample should be consistently close to the time listed in the report. It's another identifying feature that gets swept up in marketing fluff with the purity percentage. A COA that lists an exact retention time, and another test that matches that number, is more likely to be legitimate.

The peak list. A table breaking down all of the peaks visually shown in the chromatogram. This is another place to look for honesty, or a lack thereof. Ideally, this document should show the smaller impurity peaks right next to the large desired peak.

On one real BPC-157 COA, the desired peak registered at 99.82 percent on one vial and one other small impurity peak. On the second vial tested, the main peak registered at 99.99 percent and the combined percentage of the other peaks was less than 0.01 percent. On a real N-Acetyl Semax Amidate report the machine found seven total peaks, but the largest was at 99.14 percent with the other six combined being less than one percent. These are the numbers to look for in a report that proves the seller isn't hiding light impurities.

Reading the mass spec section

Mass spectral analysis has two important numbers for any peptide COA. Expected mass is the known weight of the peptide in question, and measured mass is what the machine actually weighs. A match between the two indicates the sample is what the seller says it is.

On that real Semax amidate report discussed earlier, the expected mass was 854.35 and the measured mass was 854.42. On another sample, the expected mass was 854.35 and the measured mass came out to 854.39.

Note what a match actually looks like. It is not identical numbers. Real instruments carry a small tolerance, so a couple of hundredths of a Dalton apart is a match. A whole Dalton or more apart is not, and that is the gap that matters.

Molecular weight and molecular formula are also relevant, but those are often included in the text summary beneath the graph. Look for the expected weight in Daltons (Da), sometimes listed as m/z, next to the name of the desired compound. It'll be a decimal number that should match the measured decimal weight. The 1+ and 2+ you might see to the right indicate the charge of the peptide, with the 2+ version of a peptide weighing close to double the 1+ version. Those are both valid weights, but expected versus measured is the important detail to note.

The number almost everyone misses: net content

This is something generic guides tend to forget, and it's a detail that can cost you dearly if you ignore it. Purity and content are not the same thing.

Purity is often described as the quality of the material in the vial, but what that actually means is of what substance is the vial actually composed. 24 karat gold is purer than 14 karat gold, but both are alloys. Content is also known as assay, and it's the quantity of the material in the vial. It's how much of that 24 karat gold you're actually getting.

A good way to understand the difference is to think of purity as the type of gold, and content as how much of it you're getting.

Content differs from purity in ways that can surprise even the most seasoned peptide buyer. A freeze dried vial doesn't actually contain only peptide. It has other materials in it, like residual water and counterions left from the manufacturing process. Counterions are charged particles that balance out the charge of the peptide, and common ones include acetate and chloride. Those materials all add to the weight of the vial, but none of them are peptide. And that's where HPLC can fail you, since those ions don't show up on an HPLC test with a UV detector. A good COA will actually note that injectable peptides may contain salts and sugars that weren't subtracted during the purity process.

Follow that reasoning through and it leads somewhere uncomfortable. The two numbers can disagree, and neither one catches what the other misses.

Take a vial labeled 10 mg that reports 99 percent purity and an assay of 9.0 mg. The powder is genuinely 99 percent peptide. There is just less powder in there than the label claims. You would be measuring your doses off a number that is 10 percent too high, and the purity figure would never tell you.

It can also run the other way, and that is the case people never expect. In 2024 researchers bought semaglutide from three online sellers and had it independently tested. Purity came back at 14.37, 8.97 and 7.7 percent against 99 percent advertised. But the amount of actual semaglutide in those vials was 28 to 39 percent higher than the label said, not lower. The powder was mostly other material, and there was still more drug in there than advertised. (PMID 39509151)

Both numbers were wrong, in opposite directions, at the same time. The researchers' own conclusion was that the real risk here is accidental overdose.

Purity tells you what you're getting. Content tells you how much. Ask for both.

Decoding the header and tables

Compound name. The substance tested on the COA. A Semax report will state N-Acetyl Semax Amidate, not simply Semax.

Lot or batch number. The most important cross reference on the COA. A COA is only valid for the particular batch it was created for. If your vial lists one lot number and the COA displays another, the certificate essentially has no useful information.

Date. Freshness is important. A COA made two years ago, or for an expired batch, is essentially useless.

Some labs display a results table instead of graphs, and the following comes from those tables. This is a general reference, since formats change from lab to lab.

Assay is milligrams of peptide in the vial, the quantity in milligrams.

ID by RT (Retention Time) states one of the identifying features, the time it takes for the sample to run through the column. If it's close to 1.00 it's a match for an authentic sample.

ID by Spectral is another identifying feature, and this time it's measured as a score out of 1000. A 1000 is an exact match.

Coelution control is an important detail that gets swept under the rug. It states how much of your sample is being hidden underneath other compounds sharing the same peak, and it's a reality check for any HPLC result. Scores near the top of the scale indicate what you see is a true representation of your sample.

System suitability is basically the quality control for the HPLC machine itself. Low numbers, usually less than 1 percent, indicate the machine performed well and the numbers can be relied upon.

On that real BPC-157 report, all of these matched up. ID by RT showed 1.00, ID by Spectral listed 999 to 1000, coelution control came in at 998 to 999, and system suitability listed 0.4 percent. These are the numbers to look for in an authentic report, and the details that change when a dishonest vendor uses Photoshop to alter a real COA.

A real one, annotated

Here is an actual certificate, not a description of one. Retatrutide, 10 mg, tested by an independent lab. Every number discussed above appears somewhere on this page.

123456789

1. 1The verification code. A code the lab prints so the report can be checked against the lab's own records, not the vendor's. This is the single most important thing on the page, because everything else is just pixels until you confirm the lab actually issued it.
2. 2The lot number. RETA10-0616. This has to match the number printed on your vial. A perfect COA for a different lot tells you nothing about the vial in your hand.
3. 3Identity, by mass spec. MALDI-MS confirms the molecule is what the label claims. This is the test that catches an analog being sold as the real thing, and it is the one dishonest vendors leave out.
4. 4Content: 10.49 mg. A 10 mg vial that actually holds 10.49 mg, so 4.9 percent OVER label. Legitimate, independently tested, and still not exactly 10. This is why you check assay instead of assuming.
5. 5Purity: 99.8 percent. Of the powder present, 99.8 percent is retatrutide. Note this is a different question from the line above it. Purity is what, content is how much.
6. 6Retention time: 11.077 min. How long the compound took to travel through the column. It should match the labelled peak on the chromatogram below, and it does.
7. 7One dominant peak. This is what a clean chromatogram looks like: a single tall spike at 11.077 minutes and a flat baseline everywhere else. Several large peaks would mean a mixture.
8. 8Mass: 4732.332. Compare to the stated molecular weight of 4731.2 at the top. The difference is one proton, because the instrument measures the molecule plus an added hydrogen, written [M+H]+. Expected 4732.207, measured 4732.332, off by 0.125. That is a match.
9. 9How to check it yourself. A QR code, a verification URL, and a named chemist who signed it. You can go to the lab's site, enter COA5627, and see whether the original matches the PDF you were sent.

The one thing this page cannot tell you is whether the vendor sent you powder from this lot. That is what the lot number cross-check is for, and it is the step almost nobody actually does.

A sixty second walkthrough

Now put it all together on a different report. You've opened a certificate for BPC-157, 10 milligrams.

First impressions are good. The header shows the compound is what you expect, the date is recent, and the report was prepared by an independent laboratory.

Identity is next. ID by RT shows 1.00 and ID by Spectral lists 999. Combined, those say the vial contains authentic BPC-157.

On to purity. This report lists 99.82 percent on one vial and 99.99 percent on the second vial tested.

On to content. The assay is 9.22 and 9.30 milligrams on a vial that lists 10 milligrams. That means this particular vial holds around 92 to 93 percent of the peptide stated.

All three questions answered, and this COA has passed the basic tests without a chemistry degree. Note that last number though. A perfectly legitimate, independently tested vial still came in roughly 8 percent under label. That is normal, and it is exactly why you check the assay rather than assuming.

A second one, with a problem in it

The retatrutide certificate above is clean. Most are not, and the interesting ones fail in small ways rather than obvious ones.

Here is another real report from the same independent lab, this time for BPC-157 at 10 milligrams. It is not the report described in the walkthrough just above, which came from a different lab in a table format.

Run the sixty second check on it and it passes everything. Purity 99.8 percent. Content 10.02 milligrams on a 10 milligram vial. Endotoxin passed, single clean peak, named chemist, verification code. You would close the tab satisfied.

Look for another thirty seconds and two things on the page do not line up.

123456789

1. 1A different code. Same lab as the certificate above, different report, different code. Check this one against the lab's records the same way. Having seen one real report from a lab tells you nothing about whether the second one is real.
2. 2What the methods block says. Read this line and hold on to it. It states that content was measured by HPLC quantitation, which is the correct way to do it. HPLC separates the peptide out and measures how much of it is there against a known standard.
3. 3What the results table says. The method for that same content row now reads MALDI-MS, which contradicts the line above. It matters, because MALDI is an identity test. It weighs molecules. It does not tell you how many milligrams of them are in the vial, and absolute quantities by MALDI need an internal standard rather than a bare spectrum. One of these two lines is wrong.
4. 4Content: 10.02 mg. The tightest content figure in this article, at 0.2 percent over label. Set it next to the 10.49 on the retatrutide report and the 9.22 in the walkthrough. All three are legitimate independently tested vials and not one of them is exactly 10.
5. 5Purity: 99.8 percent. Of the powder present, 99.8 percent is BPC-157. Comfortably above the 98 percent floor. Same caution as before, this answers a different question from the row directly above it.
6. 6A threshold, not a measurement. This says endotoxin came in at or under 0.05 EU/mL and passed, which tells you it cleared the limit but not by how much. Worth noticing what is absent too. The retatrutide report carried a heavy metals result and this one has no heavy metals row at all. Missing tests are easy to miss, because nothing on a page marks the gap where one should be.
7. 7The number matches the picture. 5.975 minutes stated, and the chromatogram shows one tall peak at 5.975 with a flat baseline either side. Checking the stated retention time against the graph takes two seconds and almost nobody does it.
8. 8Two identifiers, one submission. The photographed vial reads QC260899, which is the lab's own sample code and matches the header on both graphs below it. BP10-0318 in the product information is the batch. Different labels for the same sample, and the batch number is the one you cross reference against the vial a vendor actually ships you.
9. 9Mass: 1419.327. Against 1419.2 stated at the top, so a 0.127 gap and an apparent match. Do the check yourself anyway. BPC-157 is a 15 amino acid chain, GEPPPGKPADDAGLV, and adding up its residue masses gives 1419.55. The measurement is fine, since 1419.33 against a true 1419.55 sits within the tolerance this sort of reading carries. It is the reference weight printed on the certificate that is off. Note also that this peak carries no charge label, unlike the [M+H]+ on the retatrutide report, so you cannot assume the same proton arithmetic applies.

Neither problem makes this a bad vial. The results are genuinely good, and 10.02 milligrams on a 10 milligram label is the best content number on this page. What the two slips tell you is that the paperwork got less care than the testing, and that is worth knowing, because the paperwork is the only part of any of this you can personally inspect.

There is a more useful lesson buried in it. A certificate that contradicts itself in a small way is a certificate nobody proofread. Someone assembling a fake from a template makes exactly this kind of error. So does a real lab having a busy week. You cannot tell which from the page alone, and that is precisely what the verification code is for.

Third party testing versus in house

Testing is a good start, but who does the testing matters even more.

In house means the vendor tested their own product. Better than nothing, but it's still the player grading his own exam.

Third party means an outside lab did the testing, and ideally nobody but the lab itself benefits from the results.

The strongest possible option is an independent third party report that can be verified against a specific batch.

How to verify a COA is real

PDFs are digital images, and anything that's a digital image can be altered. Vendors worth trusting will note some way to verify the report against the lab's own records.

Look for a verify key or link on the COA. Some independent testing facilities print a key and a web address. Entering the key on the website will display the original report directly from the lab, not from the vendor. I've seen one real report that printed "verify" at the lab's web address with a key placed right on the page. That's the kind of detail to look for.

Match the lot or batch number to what's on your own vial. A perfectly good COA for lot 1305 is useless if your vial is lot 1420. Cross reference the date, and make sure the compound name is exactly what you purchased.

If a vendor can't provide a specific, verifiable report for a particular vial, everything listed on any PDF is just marketing.

Red flags that should stop you cold

Any one of these is enough to make you pause and reconsider.

The lot number on the COA doesn't match your vial. There may be nothing wrong with the vial, but the certificate is for a different batch and is therefore useless for your needs.

A recycled COA, one old report continually repurposed for other customers.

No independent laboratory listed, or no way to actually verify the document.

Purity without mass spectrometry and molecular weight confirmation. A clean sample is great, but the vendor has no way to prove what the sample actually is.

A cropped or screenshot report with the lab name, date, or peak summary cut off.

A suspiciously round number like 100.00 percent purity, or no peak summary at all.

A molecular weight that doesn't match the expected weight for the compound name.

Two parts of the same document disagreeing, such as a methods section naming one test and the results table naming a different one for the same row.

No content or assay listed, only the purity percentage.

Vague dates, or reference-only language tied to no specific batch number.

None of these require a degree in chemistry to spot.

The analog trap, and why mass spec saves you

This one is more complicated, and it's about the importance of molecular weight.

Take Semax as an example. When you order research grade Semax, there's a good possibility you won't be getting the original substance. There's a family of similar substances collectively referred to as analogs, including N-Acetyl Semax and N-Acetyl Semax Amidate. Semax Amidate, also called Adamax, is frequently sold alongside regular Semax as a related but different peptide. It's easy to confuse the two, and even easier to pay premium prices for the incorrect one.

Each analog has a slight modification to the structure, making them interesting research compounds in their own right, but those differences also mean the substances aren't identical to the original Semax, which has a molecular weight of 813 Da.

So how do you tell the difference between analogs? Not with an HPLC scan. Each analog has a slightly different molecular weight, but those differences are small enough to disappear into the category of light impurities on an HPLC. A mass spectrometer is the only reliable way to make the distinction, because those small differences in weight are obvious to it.

Take N-Acetyl Semax Amidate. The addition of an acetyl group puts its expected weight at 854 Da compared to regular Semax's 813 Da. If you ordered regular Semax and the weight comes back somewhere near 854, the COA has told on the product.

The name on the vial is what the vendor wants you to see. The number on the report is what you should trust.

The sixty second checklist

Run this on every COA before you trust it.

Does the compound name match what you purchased?

Does the lot or batch number match your vial?

Is the date recent, and relevant to your vial?

Is an independent laboratory listed, ideally one you recognize by name?

Does the report include a method to verify its authenticity?

Are both identity (mass spec) and purity (HPLC) listed?

Is the purity level 98 percent or higher?

Is an assay, or content, listed for the vial?

Are any of the red flags above present?

Nine items, and it only takes one reason for suspicion to find a dishonest vendor. Several reasons are even better.

My take

Reading a COA is really about learning to think for yourself. It's about demanding proof rather than accepting claims blindly, and demanding the right kind of proof at that.

Know the three questions. Know where to find the answers on a COA. And most importantly, know how to verify those answers for yourself.

Nobody is coming to do this for you. The regulatory system that would normally check any of this does not apply here, which is exactly why the document matters and exactly why so many of them are worthless.

---

This content is for educational and informational purposes only and is not medical advice. Peptides discussed are research compounds and may not be approved for human use. Nothing here should be used to diagnose, treat, cure, or prevent any disease. Always consult a qualified healthcare professional before starting any peptide, supplement, or protocol. Individual responses vary. Do not self-administer compounds without proper medical supervision.