Peptide lab results differ: why labs disagree on the same batch, and how to compare two results

Where peptide lab results differ, and where they agree

In published same-batch comparisons, the mg per vial is where labs differ most. Purity results often sit close, though not always. In a comparison Peptide Critic published on 13 May 2026, five labs tested one retatrutide batch labelled 10 mg and set their results beside the seller’s own report. All five confirmed retatrutide, and every purity result fell between 99.74% and 99.93%. The mg results ran from 10.18 to 12.80 mg, about 26% apart.

Inside each lab, results were close. No lab’s own vials were more than about 4% apart, while the labs as a group spread by 26%. That pattern points to systematic causes between labs, such as the basis of the mg or the reference standard. Peptide Critic described the mass results as showing “normal but meaningful method-driven variation”.

Published same-batch comparisons

Each row is a single published case. None of them proves that a lab is wrong. Together they show how far methods can drift apart on one batch.

Published case	Product and label	Purity	mg per vial
Peptide Critic, 13 May 2026	Retatrutide, 10 mg	99.74 to 99.93% in all six reports	Vanguard 10.18, 10.30; Analytical Formulations 10.71, 10.85; Janoshik 11.03 to 11.49 (four results, three of them from the seller’s report); TrustPointe 12.296, 12.328; Freedom Diagnostics 12.47, 12.80
Krysia’s Substack, 2 Aug 2026	Tirzepatide, 60 mg	Reported high at all three labs	BTLabs 62.3; Chromate 70.99, 71.63; Accumark 73.72, 75.44
graymarket, 3 Sep 2025	Retatrutide, one batch	Not compared	Janoshik about 30, on its vial and on nine earlier vials; Chromate 30; Krause 26 on both vials
Krysia’s Substack, 13 Aug 2026	Tesamorelin, one batch	Freedom Diagnostics 99.985%; Janoshik 97.779%	Freedom Diagnostics 10.055; Janoshik 11.057 (averages)
Same piece	Epitalon, 50 mg	Freedom Diagnostics 99.935%; Accumark 99.27%; Chromate 95.915%	Freedom Diagnostics 47.325; Chromate 55.525; Accumark 83.64 (averages)

Two points stand out. First, the same lab can sit high in one case and lower in another. Freedom Diagnostics reported the most mg in the retatrutide case and less than Janoshik on the tesamorelin batch. Chromate matched Janoshik in the graymarket case and read 18 to 19% over the label in the tirzepatide case. One comparison says little about a lab’s general direction. Second, the gaps come in two kinds. In the retatrutide case, a 26% mg gap sat beside purity results within 0.2 points. The tesamorelin and epitalon rows show purity results 2.2 and 4.0 points apart. Our report on same-batch comparisons covers each case in more depth.

Five reasons peptide lab results differ

Cause	What it moves	Example from published sources	What the report should show
Basis: net peptide, salt or whole powder	mg per vial	Peptide content of the solid 70 to 90% (Bachem)	The stated basis of the mg
Reference standard	mg per vial, every sample in one direction	Five reports wrong by the same proportion (Finnrick, May 2026)	Standard source, lot and assigned content
Salt and water	mg per vial, through the sample and the standard	TFA 21.9 to 35.2% of salt weight by formula (Erckes et al., 2025)	Counter-ion and water results, or the correction used
Single preparation and injection	mg and purity, at random	±10% check-standard acceptance at one lab (Krause)	Preparations, injections and vials behind the result, with the uncertainty
Integration	Purity	79.802% to 99.294% after a buffer peak was left out (Finnrick log, July 2025)	Chromatogram, peak table and integration settings

1. Net vs gross: what “mg” means

A label of 10 mg can mean 10 mg of peptide (net), 10 mg of the peptide salt, or 10 mg of powder (gross). The gap between them is large. Bachem, a peptide manufacturer, writes that research peptides are typically above 95% pure by HPLC, while the peptide content of the solid may range from 70 to 90%. So a gross result and a net result for the same vial can differ by 10 to 30%, with no error at either lab.

Labs describe their mg in their own words, mostly in answers to Finnrick’s method questionnaire:

• Chromate: “the recovered mass of material”; purity is not used to adjust it.
• BTLabs: “mass of vial contents × HPLC purity fraction”, described as the peptide mass found against a reference standard and corrected for purity.
• Krause Analytical: the mass recovered from the vial against reference materials, with a % of the label claim.
• Janoshik: net peptide content, without counter-ions, as it explained in the August 2026 tirzepatide piece.

Whether purity enters the mg is a difference in itself. At 99.5% purity, multiplying by purity lowers the mg by 0.5%. At 95%, it lowers it by 5%. Our guide to net peptide content shows how to convert between bases.

2. The reference standard sets the scale

Labs find the mg by comparing the sample’s detector response with a reference standard of known content, so every result inherits the content assigned to that standard. If the assigned value is 5% too high, every sample measured against it reads about 5% high (our arithmetic). Finnrick’s blog of 9 May 2026 describes one unnamed lab that issued five reports with potency wrong by the same proportion, which Finnrick put down to a likely calibration error between the standard and the samples.

Assigning that content is itself a measurement. A 2023 paper on reference standards for synthetic peptide medicines (McCarthy et al.) describes assigning content by mass balance: it subtracts impurities, acetate, TFA, residual solvents and residue on ignition, then corrects for water. The authors warn that bias in value assignment, or changes in a standard during storage, can lead to inaccurate content. In the tirzepatide piece, BTLabs pointed to method differences and the lack of USP standards. Janoshik said it standardises its own reference material by CHNS elemental analysis and NMR. Vanguard said it buys certified standards from several sources and identifies the counter-ion by ion chromatography when a client does not state it.

3. Salt and water in the powder and the standard

Counter-ions add weight that is not peptide. Erckes and colleagues (2025) calculated from the formulas that TFA makes up 21.9 to 35.2% of the weight of seven peptide TFA salts. Water adds more. Bachem notes that peptides are often hygroscopic and that water uptake reduces the overall peptide content. In the 2023 reference-standard paper, vials were opened under dry nitrogen for coulometric Karl Fischer titration, and six freeze-dried candidate lots held 1.11 to 2.79% water.

Salt and water act twice. In the sample, they decide how far a gross result sits above a net one. In the standard, they set the scale. A standard treated as pure peptide when it is a TFA salt, or one that took up water after opening, holds less peptide than assumed, so every sample measured against it reads high (our arithmetic). Our guides to TFA in peptides and water content explain both tests.

4. Single preparation, single injection

Krause Analytical states that its standard data come from a single preparation and a single injection of a single vial, and it offers a multi-vial, multi-injection option. Its check-standard acceptance of ±10% “sets the lower limit of error” at 10%, and that margin does not include vial-to-vial variation. BTLabs does no routine replicate injections and states a typical uncertainty of ±2–3%. Chromate repeats injections of the standard only and estimates 2–5%. Ichor Life Sciences uses one injection, may re-prepare or re-run a sample about 20% off its reference, and says it has not established a formal uncertainty. Each lab states its practice openly, and that is what lets a reader weigh its numbers.

Margins add up between labs. Two labs that each work within ±10% can report 9.0 and 11.0 mg for a vial that holds 10 mg, and both stay inside their own limits. A gap smaller than the two margins combined is not evidence that either lab erred. Our guide to measurement uncertainty explains how to read a ± on a report.

5. Integration: where each peak starts and ends

Purity is the main peak’s area as a share of all integrated peaks. The software draws a baseline and sets where each peak starts and ends, and the analyst decides which peaks count. Those choices move the result, sometimes by a lot. Finnrick’s public corrections log lists examples:

• 28 July 2025: “Lab A” re-analysed two tesamorelin results to leave out a second peak, identified as a histidine buffer. One purity went from 79.802% to 99.294%.
• 9 March 2026: two results were revised because a lab’s automated integration routine had picked up noise as peaks.
• 11 March 2026: a purity of 97.4% was corrected because a blank correction had been missed.

Leaving out a peak can be right, as the buffer case shows, or it can hide an impurity. Only the chromatogram and the peak table tell the two apart. The tesamorelin case in the table above, 99.985% against 97.779%, is a purity disagreement on one batch. The August 2026 piece pointed to small peaks beside the main peak and asked whether they were counted. It also noted that not every peak on a chromatogram should lower purity, and that the HPLC method affects which impurities separate. One case cannot show which number is right. Our guides on HPLC purity and 99.9% purity results go further.

Also in play: vial fill and slips

Vials differ. Chromate names the precision of the manufacturer’s filling equipment as the main cause of vial-to-vial variation. Krause expects two vials to differ, on average, by more than its analytical uncertainty. So a lab testing one vial and a lab testing another can both be right. Mistakes also happen. On 5 November 2025, Finnrick’s log records that “Lab G” entered 92 mg instead of 9.2 mg. A gap of a factor of ten is worth checking for a data-entry or dilution slip before anything else.

How to compare two peptide test results

The pattern of a gap points to its likely cause.

What you see	Likely cause	What to ask both labs
Purity within a point, mg 10 to 30% apart	Basis or reference standard	The basis of the mg; the standard’s source, lot and assigned content
One lab high or low by the same share across products	Reference standard or calibration	How the standard’s content was assigned and checked
mg close, purity several points apart	Integration, wavelength or peak rules	Chromatogram, peak table and integration settings
A gap within the two labs’ margins combined	Ordinary measurement uncertainty	Preparations, injections and vials behind each result
Results ten times apart	Data-entry or dilution slip	A recheck against the raw data
One lab’s own vials differ	Vial fill	Each vial’s result, with the mean and range

Then work through five steps:

1. Put both results on one basis. If one is gross and the other net, the gap may be the salt and water. Counter-ion and water results let you convert.
2. Compare the gap with both uncertainties. If a report prints no uncertainty, ask the lab for one.
3. Count the vials. One vial per lab cannot separate the fill from the method.
4. Ask the method questions. Finnrick’s 10-question questionnaire is a ready list: calibration, replicates, recovery, corrections, detector, how purity enters quantity, margin of error, rounding, the meaning of the reported mass and the expected vial variance.
5. Test a fair split. For a decision that matters, send vials picked at random from the same batch to both labs at the same time, and ask both for the same basis. A group test can share the cost.

What nobody publishes yet

Finnrick’s labs page says it studies “overlapping tests”, where several labs analysed samples from the same batch, across “hundreds of such comparisons”. The aim is to see whether a lab reports systematically higher or lower quantities. When we checked in October 2026, it published no per-lab offsets. In our September 2026 market research, we found no published round robin that tests one batch at many labs and reconciles net and gross results. The author of the tirzepatide piece called for that kind of round: matched samples, net peptide on one defined basis, and a named reference material with the way its potency was assigned.

How HK Analytical reports content so results can be compared

Our content per vial result is the mg of the target peptide as free base, measured against a reference standard, with the uncertainty and the % of the label claim. If no qualified reference standard exists, the value is marked “estimated”. Counter-ion and water content can go in the same order, so the share of the powder that is salt or water is a measured number. Vial-to-vial consistency gives each vial’s mg with the mean, range and relative standard deviation. Every number comes with its raw data, including the chromatograms, so another lab or a reader can check the integration. Reports are signed records and are never edited: a correction makes a new version and marks the old one superseded. Our comparison page sets what our reports print next to five other labs, and the example report shows the layout.