The single number on a peptideone vendor card is not an opinion. It is a weighted average of up to three independent testing signals, each from a different source running a different methodology. Here is what goes into that number, how it is calculated, and why one prominent source, PeptideBenchmark, is left out on purpose.
What a "composite score" means
A composite score is only as trustworthy as its inputs, and the inputs are only useful if they are genuinely independent of each other. Count the same test data twice under two different labels and the score looks more confident than it has any right to be. The peptideone Score (0–10) is built to avoid that. It blends whichever of three signals exist for a given vendor, then renormalizes the weights over whatever is present. A vendor with all three signals gets a score from three sources. A vendor with one gets a score from that source alone, and we say so.
The three sources and their weights
| Source | What it measures | Scale | Weight in composite |
|---|---|---|---|
| [Finnrick](https://www.finnrick.com) | Blind-buy safety grade (purity, dose accuracy, batch quality) | 0–10 | 45% |
| [VialAudit](https://vialaudit.com) | Blind-buy composite audit (purity, label accuracy, shipping, service) | 0–100 (rescaled to 0–10) | 30% |
| [Peptigrity](https://peptigrity.com) | HPLC purity percentage from aggregated independent lab tests | % (mapped to 0–10) | 25% |
Finnrick carries the most weight because its published methodology is the most transparent. It documents the three scoring components it uses: purity by HPLC (0–4 points), quantity accuracy against the label claim (0–4 points), and batch-information quality (0–2 points). It applies them the same way across blind-buy purchases. Finnrick converts that 0–10 internal score into A–F letter grades, plus an "X – Unrated" tier for vendors with fewer than two tests. We store and display the underlying score, attributed to Finnrick, not as our own measurement.
VialAudit adds a second audit angle. Its 0–100 score folds shipping and service signals in alongside the analytical ones (per its published methodology, purity 40%, label accuracy 25%, shipping 20%, service 15%). We rescale it to the same 0–10 range for blending. VialAudit runs its testing as blind-buy audits shipped to an alias mailbox, which makes it structurally independent of what Finnrick tests. The same vendor batch may or may not overlap.
Peptigrity publishes per-test HPLC purity data sourced from independent labs across multiple peptide purchases. We pull the mean purity percentage for each vendor from those underlying test records. We do not use Peptigrity's composite Trust Score, which blends lab purity with community reviews. Raw purity percentages bunch up near the top of the range (a vendor might reasonably land anywhere between 95% and 99.5%), so we apply a mapping that spreads the meaningful 90–100% band across the full 0–10 scale. A result below 90% maps to 0. A result of 100% maps to 10.
When more than one signal is present, the weights renormalize. If only Finnrick and VialAudit are available, their combined weight is treated as 100%, in proportion to the original 45/30 ratio.
Why PeptideBenchmark is excluded
PeptideBenchmark is a useful site, and we reference it separately in our data pipeline. But as of early 2026 its Benchmark score is itself a derived ranking that normalizes and re-blends Finnrick's vendor data with other sources. Put it into our composite next to Finnrick and the same Finnrick tests get a second vote. That is double-counting, and it inflates apparent certainty without adding anything new.
The rule is simple: a source enters the composite only if it contributes data not already represented by another input. PeptideBenchmark fails that test. We still ingest and store PeptideBenchmark scores as attributed reputation signals, and they show up on vendor profiles for reference. They are just not folded into the score calculation.
What the score does and does not tell you
The peptideone Score is a research-aggregation tool, not a purchasing recommendation. It summarizes publicly available independent testing data, attributed to the labs and sites that produced it. A higher score means independent testers have, on average, found stronger analytical results for that vendor's products. It says nothing about legality, dosing, or suitability for any purpose. Research peptides are sold for research use and are not approved for human consumption.
The full source breakdown sits next to the score on every vendor profile: which signals contributed, their individual values, and the underlying test records.