Why Two Laboratories Report Different Numbers

Send the same lot to two laboratories and the numbers will not match exactly. That is normal, and it is not evidence that one of them is wrong. Knowing how much disagreement is ordinary — and what size of gap stops being ordinary — is what makes a second opinion useful rather than confusing.

Where the differences come from

• The reference standard. Content is measured by comparing your sample against a known standard. If two laboratories use standards of different provenance, purity or assigned value, every figure that follows shifts with it. This is the single largest source of disagreement on content.
• The gradient and column. Different gradients separate impurities differently. A shoulder that resolves into its own peak on one system stays merged with the main peak on another — and a merged impurity is counted as main peak, raising the purity figure.
• Detection wavelength. 214 nm reads the peptide bond; 280 nm reads aromatic residues. The same impurity contributes differently at each.
• Integration decisions. Where a peak starts and ends, whether to split a shoulder, what counts as baseline noise. Two competent analysts can differ by a few tenths on identical raw data.
• The vial itself. Two vials from the same lot are not identical. Fill variation alone can account for several percent of content.

What size of difference is normal

As a working guide for the same lot tested twice:

• Purity within a few tenths of a percent — expected. Treat as agreement.
• Purity differing by 1–2 points — worth a question. Usually a separation or integration difference; ask to see both chromatograms.
• Purity differing by more than a few points — something substantive differs: method, sample, or lot identity.
• Content within roughly 5% — ordinary, given standards and fill variation.
• Content differing by 15% or more — not explainable by method noise. Either the standards differ materially, or the vials do.
• Identity is binary. Mass results either match the expected molecule or they do not. A genuine identity disagreement is serious and is not a tolerance question.

How to compare two reports properly

1. Same lot? Check the lot number on both, not just the product name. Most “contradictory results” turn out to be different lots. See lot numbers and traceability.
2. Same quantity being reported? Content per vial and net peptide content are different numbers; so are purity and content. See net peptide content.
3. Same method? Compare the stated method, wavelength and gradient before comparing the figures.
4. How many vials each? One vial against one vial compares two samples as much as two laboratories.
5. Are both verifiable? A number that cannot be checked at the issuing laboratory is not evidence either way. See verifying a report.

When a second test is worth paying for

Independent testing is most informative when the first report is unverifiable, when the material looks or behaves unexpectedly, when a lot is being used as a reference point for later comparisons, or when the supplier’s figures are implausibly perfect across every lot. It is least informative when run on a different lot from the one in question, which is the most common way money gets wasted on it.

If you do test one of our lots independently, send us the result either way. Published figures that do not survive an outside check are worth more to us as a correction than as a sales point — our vendor information page sets out that policy in writing.

Materials supplied by BMS Labs are for laboratory research use only and are not for human or veterinary use.