Editorial policy & independence
How a Checkthevendor grade is produced, who signs off on it, and the rules that keep the scoring honest.
A grade is our editorial opinion, grounded in disclosed, dated facts. We publish the facts we relied on and the date we gathered them, so you can check our reasoning and reach your own conclusion.
We are independent from the vendors we grade
We are not owned by, operated by, or controlled by any peptide vendor. We do not take payment to raise a grade or change a ranking.
We do earn money through affiliate links and clearly labelled sponsored placements. That is a hard line, not a soft preference: money never affects a grade or a ranking. The same published rubric applies to paying and non-paying vendors alike, sponsored slots are visibly labelled and kept out of the merit ranking, and affiliate links are disclosed where they appear. If a vendor pays us and earns a poor grade, it gets the poor grade. The full money rules live on our Disclosure page.
How a grade is produced
Every vendor is scored against eight criteria, each weighted to sum to 100%:
| Criterion | Weight | Type |
|---|---|---|
| Third-party lab testing & COA verification | 25% | measured |
| Measured purity & contaminant profile | 18% | measured |
| Transparency & disclosure | 12% | mixed |
| Reputation & customer service | 12% | editorial |
| Pricing & value | 10% | measured |
| Shipping, packaging & fulfilment | 8% | mixed |
| Payment options & checkout trust | 8% | mixed |
| Website & operational trust signals | 7% | editorial |
Each criterion is scored 0 to 100 against a published rubric, multiplied by its weight, and summed into a single 0 to 100 composite. The composite maps to a letter: A+ 97 to 100, A 93 to 96, A- 90 to 92, B+ 87 to 89, B 83 to 86, B- 80 to 82, C+ 77 to 79, C 73 to 76, C- 70 to 72, D+ 67 to 69, D 63 to 66, D- 60 to 62, and F below 60. The visual walkthrough lives on How we score.
The integrity gate
Two rules override the arithmetic, because some failures matter more than a weighted average can show:
- A vendor with no independent third-party COA cannot score above the D band, no matter how strong the rest of its profile looks.
- A vendor we find publishing fabricated or reused COAs, or fake reviews, is capped at F and flagged on its profile. We describe the specific observation we documented and the evidence behind it, and let you draw your own conclusion.
The two-step sign-off
No grade is published on one person’s say-so.
- Step one: a reviewer gathers the evidence, scores each criterion against the rubric, and writes up the reasoning with sources attached.
- Step two: a second, independent reviewer checks the evidence against the scores, confirms each measured claim is backed by a linked source, and signs off before publication.
If the two reviewers disagree, the grade is not published until the disagreement is resolved against the rubric and the evidence.
Conflict-of-interest rule
No one who has a financial tie to a vendor may score that vendor, at either step. A financial tie includes ownership, employment, paid consulting, commission beyond standard disclosed affiliate terms, or any personal stake in that vendor’s sales. Where any such tie exists, it is disclosed on the affected vendor’s page and the work is reassigned to a reviewer with no tie.
How we gather and keep evidence
Every score is tagged either measured (a verifiable fact, with the evidence linked) or editorial (clearly labelled opinion, grounded in the same disclosed facts), so you can tell the difference at a glance. Our evidence comes from three kinds of source:
- Independent test purchases. We buy as an ordinary customer would and send the product to an independent lab for analysis.
- Vendor-published COAs. When a vendor publishes a certificate of analysis, we verify it against the named lab and the stated lot before we rely on it.
- Community-reputation sampling. Named sources, a stated date range, a minimum number of data points, and exclusion of outliers and astroturfed activity.
We retain the underlying evidence and the dates it was gathered, so a grade can always be traced back to what supported it.
Confidence labels
Because evidence is not equally deep for every vendor, each grade carries a confidence label. High means at least one independent test or per-lot COA in the last 12 months plus at least 20 independent community data points. Medium means an independent COA is present but sparse, or 5 to 19 community data points. Limited means vendor-supplied data only, fewer than 5 data points, or any criterion left unscored: promising on the evidence we have, but the evidence is still thin.
How often we re-review
- Pricing and value: quarterly.
- COA and purity: per new batch, otherwise every 6 months, with an independent test-buy refresh at least annually.
- Reputation: every 6 months.
- Full re-grade: annually, or whenever the methodology changes.
The latest re-reviews are always visible on the public grade-change log.
Corrections and right of reply
We can get things wrong, and facts change. Every vendor we grade has a free right of reply, and anyone can flag an error through our corrections process. If a documented fact is wrong, we correct it and note the change. Where a vendor’s own claims are relevant, we report them as that vendor’s claims, not as our endorsement.
Who stands behind these grades
Grades are produced and signed off by the Checkthevendor editorial team under the two-step process above. The grading identity is collective on purpose: every score follows the published methodology rather than any one person’s taste, and every published grade is accountable to this policy, the corrections process, and the vendor’s right of reply.