How to Compare Peptide Lab Records Across Batches, Labs and Vendors
Education7 min readNovember 4, 2026

How to Compare Peptide Lab Records Across Batches, Labs and Vendors

A repeatable way to line up peptide COAs, third-party reports and public test records: what to log, how to match batches and methods, why purity and quantity are separate results, and a worked example from a real COA page.

To compare peptide lab records fairly, put them on the same footing before you look at any percentage. Confirm that each record is genuine and covers the batch you care about, note which lab and method produced it, then compare identity, purity and quantity as three separate results. A 99.4% from one lab and a 98.9% from another may not mean what they appear to mean, because the research peptide market has no shared report format or standard method. This guide gives a repeatable way to line up certificates of analysis (COAs), third-party reports and public test records, with a worked example from a real COA page.

It picks up where two other guides stop. For reading a single certificate, see how to verify peptide purity. For commissioning your own report, see how to send a sample to a peptide testing lab.

Four Kinds of Peptide Lab Records

Before comparing anything, sort each document by who chose the sample, because that shapes what it can tell you.

  • Vendor in-house COA. The seller ran the test and chose what to publish. Useful as routine quality control, but not independent.
  • Vendor-commissioned third-party report. An outside lab ran the test, but the vendor chose the sample and decides whether the result appears.
  • Buyer-commissioned report. A customer sent a vial they received. This is the closest thing to a look at what actually ships, though it covers one vial.
  • Public test database record. A platform publishes results by vendor and product, including results the vendor did not choose. More on these below.

A "lab tested peptides" badge can sit on top of any of these, so the label alone tells you little. Our guide to independent peptide analysis labs explains how the labs themselves differ.

Step 1: Log Every Record in the Same Fields

Records are hard to compare when they stay as separate PDFs. Copy the same fields from each one into a plain spreadsheet or notebook:

  • Compound and labeled amount
  • Vendor name and batch or lot number
  • Lab name, report number and test date
  • Who submitted the sample, if stated
  • Method details: column, detection wavelength, mobile phase, gradient
  • Identity result: expected and observed mass
  • HPLC purity percentage, and whether a chromatogram is attached
  • Quantity or net peptide content, if reported
  • Other tests: endotoxin, heavy metals, residual solvents
  • Any excipient listed, such as mannitol

Leave a field blank when the record does not give it. The blanks are often the most useful part of the comparison, because they show what a headline number is resting on.

Step 2: Confirm Each Record Is Real and About That Batch

A comparison built on the wrong document is worse than no comparison. Finnrick, a peptide testing platform, puts it bluntly in its guide to reading a COA: "Without vendor name and batch ID, a vendor-supplied COA is entirely worthless and should be ignored altogether." It also notes that a COA "does not verify what vendors sell in other batches or shipments."

Use the lab's lookup or report number where one exists, then read the header with care. A guide to verifying Janoshik reports explains that "the client name, sample name, manufacturer and batch number are entered by whoever orders the test," and that copies of one report can be bought with different header details. It describes four retatrutide reports under four company names that "share the same dates, the same vial photo and the same result." Each copy can still pass verification.

So add one more check when you log records: search your sheet for identical results under different names. A match across two sellers is a question to ask, not a second data point.

Step 3: Compare the Method Before the Number

HPLC purity is a relative figure produced by a particular method. A Peptide Journal guide to reading a COA says that without methodology details "you cannot compare results across suppliers," and that "A purity number without a method is scientifically incomplete." The details it lists are column type, detection wavelength, mobile phase, gradient and flow rate.

In practice, sort your records into three groups:

  • Same lab, same stated method. The strongest comparison. Small differences can mean something.
  • Different labs, methods stated. Compare with care. Differences in gradient, wavelength or peak integration can move a figure, so small gaps prove little.
  • Method not stated. You can record the number, but you cannot rank it against anything.

Step 4: Compare Identity, Purity and Quantity Separately

These three results answer different questions, and one cannot stand in for another.

Identity asks whether the expected compound is present, usually by mass spectrometry. It works as a gate. Finnrick notes that a failed identity could mean a different peptide, a blend, fragments or no peptide at all. Any record without an identity result is incomplete, however high its purity figure.

Purity asks how much of the peptide-related material is the target. LifeTein's explanation of peptide purity defines it as "the percentage of the target peptide relative to peptide-related impurities as measured by HPLC," commonly detected at 214 nm. Finnrick adds that the figure "does NOT measure salts, fillers, excipients, water content, endotoxins, or heavy metals."

Quantity asks how much is in the vial. LifeTein describes net peptide content as "peptide material versus non-peptide components such as moisture and counterions." Two vials can show the same HPLC purity and still hold different amounts of peptide, so never infer one from the other.

Quantity figures have their own comparison problem. Finnrick explains that a reported amount is estimated by comparing the HPLC signal with a reference standard, not by weighing the peptide. A USP multi-laboratory study on oxytocin compared an HPLC assay, quantitative NMR and amino acid analysis, and found the lowest inter-lab variability when the HPLC assay used the same peptide bulk material as the standard. For your sheet, that means the reference standard belongs in the method notes when a lab discloses it.

Comparing Batches From One Vendor

This is the most informative comparison, and also the one vendors make hardest, since many post only a current certificate.

  • Look for variation. Results across lots should be close but not identical. Peptide Journal warns that when COAs repeat across batches, "the testing was not performed individually."
  • Look for direction. Several lots drifting lower says more than one low result.
  • Look at the dates. A test date should fit the batch. Peptide Journal also flags a PDF created recently for an old batch as a sign the document may not come from archived testing.

Comparing Vendors for the Same Compound

Keep the comparison narrow: same compound, similar test dates, and the same lab where possible. Then compare how complete each record is before you compare results. A vendor showing a chromatogram, a mass result and a lot number at 98.9% has given you more to work with than one showing a bare 99.9%.

Remember selection bias too. Published records are usually the ones a seller chose to show. Our COA transparency grades cover how the vendors on this site publish their documents, and the vendor rankings put those grades alongside price and shipping.

Worked Example: Reading One Public COA Page

Sports Technology Labs publishes results on a public COA page, which makes a good practice case. Each entry shows a test date and purity figure, with an image of the certificate. At the time of writing (November 2026), logging a few peptide entries gives:

  • BPC-157: 99.43%, dated 06/10/26
  • TB-500: 99.42%, dated 03/27/25
  • KPV 5 mg: 100.00%, dated 01/31/25
  • Selank: 96.8%, dated 10/17/25
  • DSIP: 90.2%, dated 10/17/2025

The page states a minimum of "95% purity when Mannitol is required for storage stability" and "98% purity when Mannitol is not needed." Here is what the log shows:

  • Near-identical figures, different dates. BPC-157 and TB-500 differ by 0.01 points, but the tests are more than a year apart. They are different compounds, so this is no quality comparison, and the older result says less about current stock.
  • The minimum depends on a missing field. The page text does not say which products contain mannitol, so Selank at 96.8% may or may not meet the applicable minimum. That is a fair question for the vendor.
  • One result misses both minimums. DSIP at 90.2% is below 95%. Ask about the current batch before relying on it.
  • A perfect figure needs its chromatogram. A reading of 100.00% should send you to the certificate image to see the trace and method.
  • No batch history in the page text. Each product has one current entry, so a lot-to-lot comparison needs a request to the vendor. The page text does not show lot numbers, so read them from the certificate image and match them to your vial.

The page also tells buyers to contact third-party labs to confirm results, which is the right next step for any entry you plan to rely on.

Using Public Test Databases

Public databases are the closest thing to a shared record across sellers. Finnrick's homepage reported 12,729 tests at the time of writing, and its COA guide says results are organized by vendor and product, with samples labeled as public, vendor or Finnrick sourced. Record IDs can be checked at finnrick.com/verify.

Three cautions apply:

  • Check the sample source. Vendor-submitted and buyer-submitted records are not the same kind of evidence.
  • Check the sample count. Finnrick itself calls its data "one input, not absolute proof of vendor quality across all batches."
  • Read an empty page as no data. Finnrick's Sports Technology Labs page currently says "No published tests for this vendor yet." That is neither a pass nor a fail.

Common Questions

Can I compare a vendor COA with a buyer's lab report?

Yes, if both cover the same batch. Start with identity, then look at purity. If the lab or method differs, a small purity gap is expected, while a failed identity check or a clear shortfall in quantity is the serious finding.

Does it matter which peptide laboratory produced the record?

Yes. A named lab with a report number you can check is worth more than a logo on a PDF. Accreditation and a verifiable report are separate checks, and third-party testing explained covers what to ask.

All products discussed are for research purposes only. Not for human consumption.

Disclaimer: All products referenced in this article are intended for laboratory and research use only. They are not intended for human consumption. This article is for informational purposes and does not constitute medical, legal, or professional advice. This site may receive compensation through affiliate partnerships with vendors mentioned.