What 33,691 COAs Report — and What They Leave Out

This article is based on an analysis of 33,691 certificate of analysis (COA) records indexed by CertIQX between May 2001 and August 2026, drawn from 525 vendors and 33 testing laboratories.

Three findings stand out.

Mass confirmation is essentially absent. Across the entire indexed corpus, not a single COA record includes a mass confirmation result. Mass confirmation — typically via mass spectrometry — is the analytical step that verifies the substance detected in a sample is actually the compound claimed on the label, rather than something else with similar chromatographic behavior. Its complete absence from the corpus is worth attention, because purity alone does not establish identity. A certificate can report a high purity number while the underlying peak has never been positively identified as the intended compound.

Sterility and endotoxin are rarely reported. Sterility results appear on just 2.9 percent of certificates; endotoxin results on 8.2 percent. Both are meaningful for reconstituted peptide products, where the diluent and handling environment matter at least as much as the lyophilized material itself. Reports that omit them leave an evidentiary gap that no purity figure fills.

Reported purities cluster heavily at the top of the range. Of the 22,887 certificates with a purity result, 99.4 percent report purity between 95 and 100 percent, and only 0.3 percent fall below 90 percent. A distribution this concentrated invites a straightforward question: whether the underlying products are genuinely uniform, whether certificates are selectively issued for batches that passed, or whether the reporting practice itself compresses the range. This dataset cannot discriminate among those explanations — it can only document the pattern.

Two additional observations from the same corpus: certificates with a test method overwhelmingly rely on HPLC (93.9 percent of indexed records), and every indexed certificate URL in the corpus currently resolves. The latter is useful evidence that the ecosystem’s verification links have not been allowed to rot on a large scale. For those looking to verify a batch against the indexed data, CertIQX maintains the public search interface.

Limitations. This analysis describes what indexed certificate records report. It does not assess laboratory quality, confirm the contents of any physical sample, or verify the chain of custody behind any individual document. Records missing a compound, a method, or a report date were excluded from the corresponding rankings per the methodology disclosed in the CertIQX API. Sample sizes are reported above per metric. Nothing here should be read as a claim about any specific vendor, laboratory, or batch.

The pattern worth remembering is simple: a certificate is a claim about a sample, not proof of a product. The more often a category of testing is omitted, the more the omission itself becomes data.