Janoshik vs Finnrick: Comparing the Peptide Testing Labs (and Why the Lab Name Is Not the Proof)

Janoshik and Finnrick are not competitors, they are different layers of the same verification stack, and confusing them is the most common vetting mistake in this market. Janoshik is an independent laboratory that runs the analysis itself and publishes verifiable reports on its own domain. Finnrick is not a laboratory at all: it is a testing and ratings platform that states it sources samples from vendors, sends them to seven named third-party labs, and aggregates thousands of results into public vendor leaderboards. Freedom Diagnostics, the US lab in Franklin, Tennessee, is a third thing again: a high-throughput lab with a fully public accession index. All three are real, all three publish checkable records, and none of them can fix the one structural weakness that no laboratory can fix from the bench: the vendor usually chooses which vial gets submitted. A famous lab name on a PDF tells you a sample was tested. It does not tell you that the sample was representative, that it came from the batch in your box, or that it was the vendor’s worst vial rather than its best. The only thing you can actually check is whether the report’s code resolves on the lab’s own domain, whether the batch on that record matches the vial you were shipped, and how many independent results exist for that vendor rather than one.

Published 2026 - For research use only (RUO). Nothing here is medical advice, a dosing protocol, or a human-use recommendation.

In a hurry? Jump to how Pepora scores on the checklist →

---

The labs that actually exist in this market

Before comparing anything, it is worth grounding the discussion in verifiable entities. Finnrick publishes a Partner Testing Labs page listing every laboratory whose results feed its rankings, along with published test counts and date ranges. That page is the closest thing this market has to a public census of who is doing the analysis. As of the 13 July 2026 data refresh it listed seven laboratories, and their published test counts sum to exactly the 9,041 tests Finnrick reports sitewide.

Laboratory	Location	Published tests on Finnrick	Date range	Public verification portal
Krause Analytical LLC	Austin, Texas, USA	5,444	24 Apr 2025 - 1 Jul 2026	CoA Verification page on krauselabs.com
BTLabs	West Palm Beach, Florida, USA	2,020	22 Jul 2025 - 26 Jun 2026	Client payment portal; btlabtesting.com
Chromate	Hudson, New Hampshire, USA	672	17 Dec 2024 - 30 Jan 2026	Verify portal on chromate.org
TrustPointe Analytics LLC	Dorr, Michigan, USA	315	13 Feb 2025 - 8 May 2025	LIMS-served COA verification
Freedom Diagnostics	Franklin, Tennessee, USA	216	25 Apr 2025 - 13 Sep 2025	Public accession/company lookup
Janoshik Analytical	Czech Republic (no street address published)	195	14 Apr 2025 - 9 Sep 2025	Verify page + public.janoshik.com
MZ Biolabs	Tucson, Arizona, USA	179	23 Jan 2025 - 29 Sep 2025	Client portal; mzbiolabs.com

A few things fall out of that table immediately. First, Janoshik is the most talked-about name in peptide forums but is nowhere near the highest-volume lab in this dataset - Krause Analytical published more than 27 times as many tests through Finnrick. Second, several of these labs are general contract analytical laboratories with much broader books of business than peptides: Krause also lists corn and grain testing, pesticide testing and synthetic colour testing; BTLabs describes itself as an analytical contract lab covering peptides, vitamins, nutraceuticals, beverages, cell culture media and raw ingredients; MZ Biolabs describes itself as a mass spectrometry specialist serving research, medical and veterinary markets and lists DEA licence RP0584676. Third, the popularity of a lab in vendor marketing is not correlated with its analytical throughput.

One more detail worth noticing while you are in that table: Janoshik, the single most-cited lab name in this market, publishes no physical address anywhere on its public site - not on the homepage, not on the About page, and not on the reports themselves. That is not an accusation of anything. It is a reminder that the reputational weight a lab name carries in forum threads is not the same thing as the documentary record it publishes about itself.

What each one actually offers

Janoshik Analytical. Accepts samples globally and states over ten years of experience with chemical analysis of performance-enhancing compounds and related substances. Its catalogue is priced per test - a common GLP-1 class blind screen is listed at $360, a blind common anabolic screen at $145, and an hGH amount, purity and dimer analysis at $500. The commercially important feature is the blind test format: the lab is not told what the sample is supposed to be, which means identity is genuinely determined rather than confirmed. Every report carries a QR code and a unique verification key, and the site states that every report in its database is verifiable via the Verify tab or that QR code; a subset of reports is published openly at public.janoshik.com.

Freedom Diagnostics. Established 2023, Franklin, Tennessee. It states that every certificate it issues includes both HPLC purity and LC-MS identity confirmation, never one without the other, on the stated reasoning that a purity number without identity confirmation is an incomplete answer, and it advertises an average 24 to 48 hour turnaround. Its distinguishing feature is that publication is the default rather than the exception - the site’s own phrase is “public by default”: the Certificate of Analysis Lookup page lets anyone search by accession number, by Search Code, or by company name. Its live counter read 42,378 publicly verifiable certificates when we checked on 21 July 2026.

Finnrick. Not a lab. It states that it sources samples from vendors, sends them to the seven labs above, and converts the results into per-compound vendor leaderboards; it also runs a free testing route for samples sent in by customers. Its published standards are explicit and numeric. There are four: two open tiers and two members-only tiers. The Finnrick Standard requires purity of at least 99.5 percent, quantity within plus or minus 25 percent of claim, and a labelling score of at least 0.5; Finnrick Super strict, a members tier, requires 99.8 percent purity, plus or minus 15 percent quantity, and a labelling score of at least 2.0. Two “purity focus” modes relax quantity to plus or minus 35 percent and drop the labelling requirement entirely. Results are time-weighted with a 730.5-day half-life, so a two-year-old test counts about half as much as a fresh one, and thin test histories are pulled toward a peer baseline so a single lucky result cannot inflate a rating. Finnrick also states plainly that it does not rewrite, remove or soften results because a vendor objects, and that it does not accept payment to change rankings.

Chromate deserves a specific mention because it runs what it calls the Chromate Testing Project: unannounced testing of products the manufacturer did not submit, which the lab frames as “keeping manufacturers honest”. That is the only model in this list that structurally defeats vendor selection.

Gameability: five ways a real lab name still tells you nothing

This is the part that matters, and it applies to every lab above equally. None of what follows is an accusation against any laboratory. The labs are doing analytical chemistry on the vial they receive. The weaknesses are in the submission and publication layer, which is controlled by whoever pays for the test.

1. Sample selection is the vendor’s, not the lab’s. A laboratory reports on what arrives in the envelope. If a vendor tests ten vials and submits the best one, the resulting COA is technically accurate and epistemically worthless. This is textbook selection bias, and it is invisible on the certificate: nothing on a report distinguishes a randomly drawn vial from a hand-picked one. Finnrick’s own methodology page is admirably direct that it “sources samples from vendors” for much of its dataset, which means the aggregator inherits the same bias - it just dilutes it across many more samples and many more time points. Only unannounced purchase-and-test, like Chromate’s project or samples submitted by buyers rather than sellers, removes the vendor’s hand from sample choice.

2. Publication can be opt-in. A verifiable report and a published report are different things. Janoshik’s contact form carries “Public Report Request” as a named subject option, which tells you plainly that reports are private by default and become public when the submitter asks. That is a legitimate confidentiality policy, but it means the public database is a self-selected subset: you see the reports someone wanted you to see. Freedom Diagnostics takes the opposite approach and publishes by default, which is why its index is orders of magnitude larger. When you evaluate a vendor, the useful question is not “does a good report exist” but “does a company-name search on the lab’s index return the vendor’s full set, including the unflattering ones.” Our analysis of what independent testing really shows works through why this makes public databases useless for estimating market-wide failure rates.

3. One vial is not one batch, and one batch is not the product line. Purity is batch-specific. A certificate documents the lot the lab sampled, at the date it sampled it. Nothing carries that number forward to the next synthesis run, and nothing carries it sideways to a different SKU. This is why the single highest-value check is the least glamorous: does the batch or lot number on the COA match the batch printed on the vial you were actually shipped? Finnrick formalises this with a labelling score from 0 to 2 that awards full marks only when the batch identifier is printed on the product container itself, half marks when it is on a separate physical support shipped with the container, and zero when it is website-only or absent - because a website claim detaches from the product the moment it leaves the warehouse.

4. Scope creep in the reader’s head. A purity COA measures purity. It does not measure fill quantity unless quantity was ordered, and it does not measure sterility, endotoxin, or residual solvents unless those specific assays were requested. Readers routinely treat “99.6% HPLC” as a general safety statement. It is not. Published market-surveillance work on peptide vials test-purchased from online sellers without a prescription requirement makes the point concretely: samples in that study were free of viable microorganisms at the time of testing and still carried detectable endotoxin in every vial, while measured purity landed far below the at-least-99-percent figure on the labels. Identity, purity, fill and contamination are four separate questions, and a COA answers only the ones it was paid to answer.

5. The PDF is the weakest artefact in the chain. A certificate is a document, and documents are editable. Every meaningful protection is on the lab’s side of the fence: Janoshik’s QR code and unique key, Freedom’s accession and Search Code, Krause’s CoA Verification page, Chromate’s Verify portal, TrustPointe’s LIMS-served certificates, which the lab says exist specifically for “verification and protection from alteration”. The verification is only worth something if you navigate to the lab’s domain yourself rather than following a link the vendor provides. Our companion walkthrough at how to verify a peptide COA by accession lookup covers the mechanics step by step, and how to read a peptide COA covers what the fields actually mean once the record resolves.

A sixth, subtler problem: labs disagree with each other

Even with an honest sample and an honest lab, purity is a method-dependent measurement, not a physical constant. Finnrick sent its partner labs a ten-question methodology questionnaire covering quantification basis, whether replicate injections are averaged, whether peptide recovery on reconstitution is assumed or measured, which detector is authoritative for quantity, stated margin of error, and whether reported values are rounded or raw. Those questions exist because the answers differ from lab to lab and materially move the number.

As of 21 July 2026, three of the seven partner labs had published questionnaire responses: BTLabs and Chromate answer on their own Finnrick lab pages, and Krause Analytical’s response is published as a scanned document dated 27 January 2026 and linked from the Partner Testing Labs page. The other four - TrustPointe, Freedom Diagnostics, Janoshik and MZ Biolabs - had no published answers. That is a transparency gap worth knowing about, and it is why Finnrick estimates inter-lab bias statistically from overlapping tests rather than declaring any single lab correct.

The practical rule: treat sub-one-percentage-point differences between labs as method noise. Treat order-of-magnitude differences as a product problem.

What this means for vetting a vendor

Strip the brand names out and the checkable criteria are short. Every one of these can be verified before you spend money.

Check	How you verify it	Pass condition
Named third-party lab	The lab is identified on the certificate and has its own live domain	You can reach the lab’s site independently
Resolvable identifier	Type the lab’s domain yourself, enter the accession/report code	Code returns a record matching the PDF’s compound and purity
Batch match	Compare lot on the record to the lot printed on the vial	Identical, and the batch ID is on the container not just the website
Both purity and identity	Report shows HPLC purity and mass spec identity confirmation	Both present; purity alone is incomplete
Full set, not one report	Search the lab’s index by the vendor’s company name	Multiple reports return, including recent ones
Coverage honesty	Check whether a report exists for the exact SKU you want	Vendor does not imply catalogue-wide testing from one SKU
Independent corroboration	Look the vendor up on an aggregator with published standards	Multiple samples over time, not a single data point

Two additional notes on honest framing. First, no laboratory in this comparison advertises ISO/IEC 17025 accreditation for peptide assays on its public site as of July 2026. That is a real limitation of the whole sector, not a knock on any one lab, and anyone claiming a specific lab is “the accredited one” should be asked for the certificate number and the accreditation body. Second, an aggregator rating is a summary of published evidence, not an endorsement - Finnrick states on its methodology page that its ranking “is not an endorsement or medical recommendation”.

If you want the wider vendor-level checklist rather than the lab-level one, how to vet a US peptide vendor and the vendor transparency scorecard extend this into shipping, sourcing and disclosure. The peptide vendor red flags guide covers the failure patterns. And once you have a verified vial in hand, the reconstitution calculator, cost per mg calculator and syringe units calculator handle the arithmetic.

How Pepora scores

Applying the criteria above to Pepora (peporalabs.com, ships from the US) gives a split result, and the split is the honest part.

Pepora publishes Freedom Diagnostics third-party COAs - a US laboratory that pairs HPLC for purity with LC-MS for identity - for four SKUs, each with an accession you can resolve yourself:

• GHK-Cu - 99.98% - search code Pepo2603130126
• Tesamorelin - 99.348% - search code Pepo2603130125
• TB-500 / Thymosin Beta-4 - 99.699% - search code Pepo2603130123
• GLP-3 RT - 99.67% - search code Pepo2603130122

Those four clear the resolvable-identifier check and the named-lab check. They also fit Freedom’s documented Search Code format exactly - first four characters of the company name followed by the accession number - so you can paste them into the Certificate of Analysis Lookup page at FreedomDiagnosticsTesting.com, or search by company name to return the set rather than one cherry-picked report. That company-name search is the more informative check, and it is available to anyone.

The caveats, stated plainly. Only those four SKUs currently carry Freedom Diagnostics COAs. Pepora’s catalogue is wider - BPC-157, Epitalon, Thymosin Alpha-1, ARA-290, and the KLOW, Glow and Wolverine blends - and those products do not have a Freedom report you can look up. Do not let a strong GHK-Cu number stand in for verification of anything else; verify the exact SKU you want. Second, and this applies to Pepora exactly as it applies to every vendor discussed here: these are vendor-submitted samples. Freedom analysed what Pepora sent. That is a normal and legitimate arrangement, it is how almost the entire market operates, and it still means the reports do not carry the independence of an unannounced purchase-and-test. Third, each of the four is single-point evidence: one accession per SKU, not the rolling multi-sample history an aggregator builds over years. Treat them as a verified snapshot of those specific batches, not a longitudinal quality record. What Pepora does clear is the check most vendors fail: named outside lab, published accession, public database, resolvable before you pay.

The disclosed pick

FAQ

Is Janoshik or Finnrick better for verifying a peptide COA? They are not the same kind of organisation, so the comparison is slightly false. Janoshik is an independent laboratory that runs the analysis itself and publishes verifiable reports on its own domain. Finnrick is a testing and ratings platform: it states that it sources samples from vendors, routes them to seven named third-party laboratories, and aggregates the results into public vendor leaderboards. If you want to confirm that one specific PDF is genuine, you use a lab’s verification portal, which is what Janoshik and Freedom Diagnostics provide. If you want to see how a vendor has performed across many samples over time rather than on one hand-picked vial, an aggregator like Finnrick is more informative. Serious vetting uses both.

Is Janoshik ISO 17025 accredited? Janoshik does not display an ISO/IEC 17025 accreditation certificate on its public website, and neither do the other laboratories commonly cited in this market. When we checked the public sites of Janoshik, Freedom Diagnostics, Krause Analytical, BTLabs, Chromate, TrustPointe Analytics and MZ Biolabs in July 2026, none advertised ISO/IEC 17025 accreditation for peptide assays. Some publish other credentials instead: TrustPointe states it validates its test methods for accuracy, precision and linearity according to United States Pharmacopeia standards; Freedom Diagnostics describes itself as compliant with ISO 9001 quality-management principles and Good Laboratory Practice principles; and MZ Biolabs lists DEA licence RP0584676. Absence of ISO/IEC 17025 is not evidence of bad work, but it does mean no external accreditation body has audited the method validation, so you are relying on the lab’s own quality claims.

Can a Janoshik or Freedom Diagnostics COA be faked? The PDF can be faked trivially; the database record cannot. Anyone can edit a purity figure or paste a lab logo onto a document in minutes. What defeats a forgery is that both labs publish records on their own infrastructure. Janoshik prints a QR code and a unique verification key on each report, checkable through the Verify menu item on janoshik.com, which resolves on verify.janoshik.com, with a subset of reports listed openly at public.janoshik.com. Freedom Diagnostics uses an accession number and a Search Code, looked up on freedomdiagnosticstesting.com. If you type the lab’s domain yourself rather than clicking a vendor’s link, and the code returns a matching record, the report is real. If it returns nothing, the document is unverified regardless of how official it looks.

What is a Freedom Diagnostics Search Code and how do I use it? Freedom Diagnostics documents the format on its Certificate of Analysis Lookup page: the Search Code is the first four characters of the company name followed by the accession number, it appears in a field labelled Search Code in the top right table of the certificate, and codes are not case sensitive. So a report issued to a company beginning Pepo with accession 2603130126 has the Search Code Pepo2603130126. You can also search by full company name to return every certificate associated with that vendor, which is more useful for vetting because it shows you the whole set rather than the one report a seller chose to show you. The lab’s live counter showed 42,378 publicly verifiable certificates when we checked on 21 July 2026.

Why do two labs report different purity for the same batch? Because purity and quantity are method-dependent, not absolute physical constants. Different laboratories use different calibration bases, sample preparation, replicate handling, rounding conventions and detectors, and they differ in whether the authoritative quantitation signal is UV or mass spectrometry. Finnrick’s Partner Testing Labs page states that it does not assume any single lab represents ground truth, and that it instead estimates systematic inter-lab bias empirically by analysing large numbers of overlapping tests where multiple labs analyse samples from the same batch. Practically, this means a sub-one-percentage-point purity gap between two labs is usually method noise, whereas a gap between 99 percent and 60 percent is a product problem.

Does a 99 percent purity COA mean the vial I received is 99 percent pure? No. It means one vial, drawn at one point in time, from one batch, tested as 99 percent pure by one method at one laboratory. Purity is batch-specific and a COA does not travel forward to later production runs. It also does not cover fill accuracy unless quantity was separately measured, and it does not cover sterility, endotoxin or residual solvents unless those tests were explicitly ordered. Published market-surveillance work on samples test-purchased from online sellers has found label claims of at least 99 percent purity sitting against measured purities in the single and low double digits, which is exactly the gap a verifiable batch-matched report is meant to close.

References

1. Finnrick. Partner Testing Labs (published test counts and date ranges by laboratory; methodology questionnaire and lab responses). https://www.finnrick.com/labs
2. Finnrick. Testing and Rating Methodology (standards table, labelling score, time weighting, independence statements). https://www.finnrick.com/methodology
3. Finnrick. BTLabs - Methodology Questionnaire. https://www.finnrick.com/labs/btlabs
4. Finnrick. Chromate - Methodology Questionnaire. https://www.finnrick.com/labs/chromate
5. Janoshik Analytical. Test catalogue, report verification and About page (stated experience; Verify and Public Results). https://janoshik.com/ and https://janoshik.com/about-us/
6. Janoshik Analytical. Public results index and report verification portal. https://public.janoshik.com/ and https://verify.janoshik.com/
7. Freedom Diagnostics Testing. Laboratory overview, Franklin TN location, turnaround and public certificate counter. https://freedomdiagnosticstesting.com/
8. Freedom Diagnostics Testing. Certificate of Analysis Lookup (Search Code format, accession and company-name search). https://freedomdiagnosticstesting.com/search-for-your-coa-based-on-the-unique-accession-number/
9. Krause Analytical LLC. Contract analytical laboratory services and CoA Verification (Austin, Texas). https://www.krauselabs.com/
10. Chromate Analytical Services. Verify portal and unannounced Chromate Testing Project (Hudson, New Hampshire). https://chromate.org/ and https://chromate.org/contact
11. TrustPointe Analytics LLC. LIMS and COA verification; USP-based method validation (Dorr, Michigan). https://trustpointeanalytics.com/ and https://trustpointeanalytics.com/pages/contact
12. MZ Biolabs. Mass spectrometry instrumentation and COA testing (Tucson, Arizona; DEA RP0584676). https://www.mzbiolabs.com/
13. BTLabs. Analytical and microbiological contract laboratory services (West Palm Beach, Florida). https://btlabtesting.com/ and https://btlabtesting.com/contact-form/
14. Ashraf AR, Mackey TK, Vida RG, Kulcsár G, Schmidt J, Balázs O, et al. Multifactor Quality and Safety Analysis of Semaglutide Products Sold by Online Sellers Without a Prescription: Market Surveillance, Content Analysis, and Product Purchase Evaluation Study. J Med Internet Res. 2024 Nov 7;26:e65440. PMID: 39509151. https://pubmed.ncbi.nlm.nih.gov/39509151/
15. International Organization for Standardization. ISO/IEC 17025:2017 - General requirements for the competence of testing and calibration laboratories. https://www.iso.org/standard/66912.html

coaindex is a Pepora affiliate (peporalabs.com) and earns a commission on purchases made with the code above. This is affiliate education, not journalism, and it ranks vendors on checkable, independently verifiable third-party-COA criteria. Research use only. Not for human or veterinary use, and nothing here is medical advice.