What we did
A full-database analysis of 224 independent tests across 20 companies and 55 compounds.
Methodology How these samples were obtained and tested
- The cohort is the complete Institute testing database as of publication, with no exclusions.
- Companies are scored using Index Score™ methodology v3.0, six weighted dimensions, with Purity & Testing computed directly from laboratory results.
- Quartiles are computed on current Index Score™ and recomputed at each publication, so the comparison tracks the market rather than a fixed panel.
- Purity averages are unweighted means across every sample with a measured purity result.
- No company was given advance sight of the aggregate findings, though all had prior notice of their individual results.
| Company | Index Score™ | Samples | Avg. purity | Verified |
|---|---|---|---|---|
| Sovrn Research | 98 | Pending | Pending | Pending |
| American Peptides | 94 | Pending | Pending | Pending |
| Oath Research | 92 | Pending | Pending | Pending |
| Apex Peptides | 91 | 19 | 98.2% | 19/19 |
| Research Peptides Co. | 90 | 16 | 97.7% | 16/16 |
A 5.1-point purity gap inside one market
Top-quartile companies average 97.9% measured purity. Bottom-quartile companies average 92.8%. Same compounds, same channel, overlapping prices.
Under-filling is the failure nobody checks for
A purity-only certificate of analysis passes an under-filled vial cleanly. Quantity verification requires a separate assay that almost no seller commissions and almost no buyer requests.
This is the Institute's annual assessment of the market we cover: 20 companies, 55 compounds with testing evidence, and 224 independent laboratory tests accumulated since we began publishing. It is not a market-size report and contains no revenue projections. It is an assessment of whether the things being sold are what they claim to be.
The short answer is that 92% of the time, they are. 205 of 224 samples met their label claims on both purity and content. 11 showed measurable discrepancies, and 8 failed verification outright. For an industry with essentially no mandatory quality regime, that is a better baseline than we expected when we started, and a worse one than any buyer should be comfortable with.
The more important finding is about distribution rather than average. The top quartile of companies by Index Score™ averaged 97.9% measured purity. The bottom quartile averaged 92.8%. That is a 5.1-point gap between two sets of businesses selling the same compounds, into the same market, at broadly overlapping prices.
That gap has widened over the two years we have been measuring it, and the mechanism appears straightforward. Companies at the top have invested in documentation, named manufacturing partners and batch-level testing, largely because a segment of their customers began asking for it. Companies at the bottom have not, because their customers do not ask. There is no regulatory floor pushing the bottom up, so the bottom stays where it is while the top pulls away.
Three structural observations follow from the data. First, transparency remains the most reliable single predictor of test performance we have found, more predictive than price, company age, or catalogue size. Second, quantity accuracy is a quieter failure mode than purity and gets far less attention; under-filled vials pass a purity-only COA cleanly. Third, the compounds attracting the most demand relative to their evidence base, the investigational GLP-1s, the newer repair peptides, are consistently where we find the widest quality spread.
On the regulatory picture: the research-chemical channel continues to operate in the space between "not sold for human consumption" as a legal formulation and the obvious reality of who is buying. We do not expect that to resolve soon. In the absence of a regulator setting a floor, independent testing is the only mechanism that distinguishes companies doing the work from companies claiming to, which is the entire reason this organisation exists.
Our commitment for the coming year is more of the same, at greater volume: more compounds under test, more samples per company, published score histories so movement is visible, and every failed result published in full regardless of who it embarrasses.
