The 11 signals we score every peptide vendor on

Most peptide "best-of" lists are lightly-reworded affiliate pitches. Ours is not. Every vendor we audit is scored on the same 11 signals on a 0-to-100 scale per signal, and the totals roll up into one of three verdicts: routable, conditional, or hold. The same rubric runs against an Rx telehealth pharmacy, a Bangkok JCI hospital, a Singapore aesthetic clinic, a US 503B compounder, and a research-chemical seller · we do not lower the bar by compound class or channel.

This post is the explainer. The same 11 signals, with what a low (under 50), mid (60 to 75), and high (80 and up) score looks like in practice. The full per-vendor breakdown lives at /vendor; the methodology overview with worked examples is at /methodology.

1. COA on every lot

Certificate of Analysis per batch from a third-party lab. Identity + purity confirmed independently of the seller.

• Low (under 50): No COA published, or only first-party self-reported testing, or COAs that independent reviewers have flagged as missing verification codes / standard client fields.
• Mid (60 to 75): Self-reported testing, OR third-party testing where the lab is not always disclosed, OR third-party COAs that lag the lots being sold.
• High (80 and up): Third-party HPLC + LC-MS per batch, current (analysis date within 90 days of production), reputable lab named (Janoshik, Anresco, MZ Biolabs, or equivalent).

2. Cold-chain verified

Documented temperature handling for compounds requiring 2 to 8 °C storage. Includes shipping, the dispensing pharmacy fridge, and the transit gap.

• Low: No cold-chain language, or generic claims with no documentation. USPS/standard-mail shipping for heat-sensitive compounds.
• Mid: Method named (insulated pouch + cold pack), differentiated by compound class. No active temperature monitoring.
• High: JCI-accredited hospital pharmacy or 503B sterile facility, validated cold chain end to end, temperature-logged shipping.

3. Rx legality

Whether the channel includes a prescription layer enforced by a regulator. This is the structural ceiling for research-chemical sellers · the channel cannot offer what an Rx channel offers, regardless of how good the COA is.

• Low: Research-chemical sale model with no Rx layer. Score floors near zero by definition.
• Mid: Single-jurisdiction Rx with limited cross-border posture, or a 503A pharmacy with active regulatory enforcement weight (FDA warning letter, state inspection finding, branded-Rx lawsuit).
• High: Multi-state-licensed Rx pharmacy or a JCI hospital with established Rx infrastructure. Active license, current registrations, no enforcement overhang.

4. Compound identity

Independent confirmation that the labeled compound is actually what is in the vial. Distinct from COA · COA is the document, compound-identity is the molecule-level verification.

• Low: No verification, or known counterfeit risk in the channel.
• Mid: Self-reported assay results, or third-party verification with inconsistent batch coverage.
• High: Janoshik (or equivalent) third-party identity verification per batch, mass-spec confirmation, peer-spot-checked by community over time.

5. Dose accuracy

Concentration claims match assay results within tolerance.

• Low: Documented dosing inconsistencies in the COA history, or the channel does not enable the user to verify dose claims.
• Mid: Spot-check verification consistent, no published inconsistencies, but no formal dose-accuracy program.
• High: 503B-grade FDA-inspectable compounding, dose-accuracy as a published QA metric.

6. Endotoxin testing

Sterile-injectable safety signal. Endotoxin is what hurts you fast if a sterile-process gap escapes.

• Low: Endotoxin testing not published; research-chem channels rarely run LAL endotoxin per batch.
• Mid: 503A pharmacy compliance · USP <797> sterile compounding standards, but lower bar than 503B outsourcing facilities.
• High: 503B + LAL endotoxin testing published per batch, FDA-inspected facility.

7. Refund posture

What happens when a vial arrives broken, the lot is dud, the order ships wrong, or the vendor goes offline mid-cycle.

• Low: No published policy, or policy that shifts all blame to the buyer, or operational halt that ended fulfilment.
• Mid: Published policy with reasonable conditions, but case-by-case discretion in practice.
• High: Written policy + real track record of honoring it (community reviews corroborate) + generous on edge cases.

8. Channel clarity

How honestly the vendor labels what it actually is. A clinic that says "clinic," an Rx pharmacy that says "Rx," and a research-chem seller that says "research-only" all score high. Aesthetic-spa front for a clinic-grade flow, or "wellness" framing on a research-chem channel, scores mid or low.

• Low: Channel hidden behind soft language, or implied prescription where none exists.
• Mid: Mostly clear positioning with some marketing-overreach language.
• High: The vendor describes its own channel accurately and consistently · including what it cannot do.

9. Support quality

Pre-sale inquiry handled within 24 hours by a named human. Post-sale support handled within 48 hours.

• Low: No reply within 72 hours, template-only replies, or ghost-after-payment patterns.
• Mid: Reply within 24 to 48 hours with substantive content. Real humans, not great responsiveness.
• High: Reply within 12 hours, substantive, willing to discuss product specifics. Often international-patient ops at a JCI hospital or a dedicated concierge layer.

10. Price transparency

Whether the price is published before you engage.

• Low: "Pricing on request" or quote-only with no public anchor. Hidden behind a consult-fee wall.
• Mid: Public pricing on the public site, but only for some products or only after some clicks.
• High: Per-mg or per-month price published on every PDP, with full per-dose-step breakdown when relevant.

11. Longitudinal retention

Whether the vendor sustains a returning patient base over time. Years in operation, named clinical staff continuity, repeat-customer evidence in community discussion.

• Low: Halted operations, recent rebrand under reputational pressure, or under 12 months of trading.
• Mid: 1 to 3 years stable, named team, no major rebrand.
• High: 3+ years stable, consistent quality across reviews over time, no major rebrand or fire-sale event.

How the verdicts roll up

The 11 signal scores feed into a final score (0 to 100) and one of three verdicts:

• Routable · the matchmaker actively recommends this vendor for users in the relevant region. The visit-website CTA is the primary action on the scorecard.
• Conditional · the vendor exists, the documentation exists in some form, and the user gets the full per-signal context before deciding. Active regulatory action, supply intermittency, or a structural rx-legality ceiling typically lands a vendor here.
• Hold · the vendor is offline, under active enforcement that halted operations, or our audit caught something that means we will not route a user there until the situation changes. The visit-website CTA is replaced with a "Not routable · on hold" pill. Scorecard stays up because the page is the answer to "what happened to vendor X." See What 'on hold' means for the editorial reasoning.

Sub-routable triggers

Seven specific events cap a vendor below the routable line and attach a mandatory risk disclosure:

1. FDA warning letter, DEA action, or documented regulatory enforcement 2. Criminal conviction tied to the vendor or named principals (non-publication for legal reasons) 3. Verified counterfeit or mislabeled product, lab-confirmed (non-publication) 4. Permanent closure (site dead 90+ days + no owner response) 5. Systematic fake-review patterns, forensically confirmed 6. Vendor refused COA request more than twice without credible reason 7. Vendor ghosts payment for 30+ days and no refund

Two of those (criminal conviction, lab-verified counterfeit) result in non-publication for defamation reasons. The other five surface with disclosure · the vendor still gets a public scorecard, just with the trigger documented.

Why we publish this

Because the alternative is "trust us." We do not want to be trusted on our word. We want to be trusted on our rubric. If you disagree with a signal weighting, email partner@panya.health · we re-score on evidence, not on volume. Pay-to-rank does not exist here.

If you want our match

Take the quiz. The matchmaker routes by your goal, region, and budget signal, against the same 11 signals scored on every public scorecard. The vendor we recommend has the rubric breakdown linked from its card.