The method behind every verdict. What we count, who reviews, and how often we change our minds.
Why the verdict system exists
If you’ve spent any time researching a peptide, you know the experience: you find a study that looks promising, then a forum post that calls it dangerous, then a podcast that calls it miraculous, then a regulatory page that says it’s not approved, then nothing useful at all. After two hours, you know less than when you started.
We wrote Peptide Repo to fix that — not by simplifying the science, but by structuring it. Every peptide we cover gets one of five verdicts. The verdict is the answer to a single question:
“Based on the evidence available right now, what should a reasonable person do with this peptide?”
The verdict is not the article. The article is still long, still nuanced, still cites the studies. But the verdict is what you can carry with you. It’s a decision-shaped summary of a long-form review, designed to be honest about uncertainty, not to flatten it.
The five verdict states
These five states are the entire taxonomy. We do not have stars, scores, or grades. We have these.
✓ Established
Strong human evidence — typically multiple randomized controlled trials in the relevant population, plus regulatory approval in at least one major jurisdiction for the indication being discussed.
A peptide that’s Established for one indication may be Investigational for another. Semaglutide is Established for type 2 diabetes and obesity, Investigational for cardiovascular outcomes outside those populations, and Insufficient Evidence for general “longevity” use. The verdict is always indication-specific.
◐ Promising
Meaningful human data exists — typically multiple human trials, but not yet at the scale, quality, or replication of Established. Mechanism is well-characterized. Reasonable clinicians prescribe or recommend it for some patients.
This is the verdict for compounds that are likely on their way up, but where the prudent answer is still “this is moving, don’t bet your protocol on it yet.”
? Investigational
Mechanism is plausible. Animal data may be strong. Human trials are limited — typically a handful of small trials, often single-arm, often in non-representative populations. Many things in this category will graduate to Promising in the next decade. Many will not.
This is where most of the peptides in popular discourse live. Investigational is not a dismissal. It is the truthful posture for a substance that interesting people are paying attention to but for which the evidence has not yet caught up.
⊘ Insufficient evidence
Not enough data to say anything useful. The peptide may have appeared in case reports or in a few in vitro studies. Animal evidence may be inconsistent. Human data is essentially absent.
This is the verdict that most peptide guides refuse to give, because it doesn’t sell anything. We give it freely. If a peptide should not be in your protocol because nothing is known about whether it works in humans, the most useful thing we can do is name that.
⚠ Cautionary
There is an active safety signal worth naming, or a pattern of misuse that has caused harm. Cautionary is not “I disagree with this peptide’s vibe.” It is reserved for compounds where the evidence shows a meaningful risk, not just an absence of efficacy.
A peptide can be Cautionary and Promising simultaneously — strong efficacy data with a real adverse event profile. We say so, and we name both.
What evidence moves a verdict
Verdicts are not opinions. They follow from a structured read of the evidence base. The hierarchy we use, in order of weight:
- Multi-site RCTs in the indication being discussed, ideally with active comparators
- Single-site RCTs with adequate power and pre-registered endpoints
- Open-label and single-arm trials in human populations, with attention to selection effects
- Mechanistic and pharmacokinetic studies in humans
- Animal model studies (which are necessary but never sufficient for a Promising verdict or higher)
- Case reports and case series
- In vitro and computational evidence
A Promising verdict requires meaningful evidence at level 3 or higher. Established requires level 1 or 2. Investigational can rest on level 4–5. Insufficient evidence is what we use when the evidence above level 5 is sparse or absent.
We do not weight evidence by recency. A well-designed trial from 2010 is not made obsolete by a small open-label paper from 2024. We weight by quality.
We do not weight by author affiliation, by journal prestige, or by whether the result agrees with what we’d hope to find. We weight by what would convince a careful clinician.
Forum reports, influencer reports, and personal anecdotes are not evidence. They sometimes prompt us to revisit a peptide, but they do not move a verdict by themselves. If a peptide is suddenly trending, we go back to the journals.
How a monograph gets written and reviewed
The current process, as of this writing:
- Topic selection. A peptide is added to the queue if it meets the criteria on the About page.
- Literature pull. A first-pass search across PubMed, ClinicalTrials.gov, the EU Clinical Trials Register, and any active regulatory filings. Every paper that meets relevance criteria goes into a per-peptide reference table.
- First draft. Written from the reference table, working from the highest-quality evidence down. Verdict assigned at the end of the writing pass, not the beginning — we want the writing to surface the verdict, not the other way around.
- PR Editorial review. The draft is reviewed by PR Editorial. The reviewer’s job is to challenge the verdict and the supporting prose, not to confirm them. If the review uncovers evidence that wasn’t in the first draft, the verdict is reconsidered before publication.
- PR Clinical Review. The draft is reviewed by PR Clinical Review for clinical-accuracy issues — dose ranges against current label, contraindications, drug interactions, indication wording precision, adverse-event language accuracy, regulatory-claim specificity. Clinical Review’s pass is the final gate before publication.
- Publication. With a date, a verdict, and a version number.
- Re-review. See below.
We do not run pre-publication reviews past the peptide’s manufacturer, partner, or distributor. They will see the published version when everyone else does.
How often verdicts are re-reviewed
Every monograph carries a “next-review-by” date in the page metadata. The default cadences:
- Established: every 24 months
- Promising: every 12 months
- Investigational: every 12 months
- Insufficient evidence: every 18 months (low priority — the field tends to stay quiet)
- Cautionary: every 6 months (high priority — safety signals evolve fast)
A verdict can be re-reviewed earlier than the scheduled date if a major trial publishes, a regulatory action lands, or a credible adverse-event pattern emerges.
When a verdict changes, the previous verdict is preserved in a “Verdict history” section at the bottom of the monograph, with the date of change and the evidence that prompted it. Verdicts do not disappear silently.
What this method can’t do
We’re not blind to what verdict-driven journalism can’t capture.
It can’t tell you what’s right for you. A Promising verdict for a peptide does not mean it’s right for any particular person. Drug-drug interactions, individual physiology, and the realities of self-administration are not captured in our verdict. They are captured in a clinician relationship, which is the layer above us.
It can’t substitute for medical judgment. “Reasonable clinicians prescribe this” is not “you should take this.” A verdict tells you the state of the field, not the state of your body.
It can’t anticipate every misuse. A peptide that is Established for hospital-administered IV use can be dangerous when self-injected at home. We try to flag delivery-route caveats, but we are not running every reader’s protocol.
It will sometimes be wrong. The literature changes. New trials publish. Verdicts move. The corrections log on every monograph is the receipt for that.
Worked example: how we’d verdict BPC-157
To make this concrete, here’s the abbreviated reasoning behind a real verdict on the site.
Mechanism: Plausible. Animal models show angiogenic and cytoprotective effects.
Animal evidence: Strong, including consistent results in tissue-repair models.
Human evidence: A small number of trials, mostly open-label, mostly in dental and orthopedic contexts. None with the scale, blinding, or replication that would justify Promising.
Regulatory status: Not approved as a medicine in the US, EU, or most major markets. Available in some jurisdictions through compounding pharmacies under specific clinical contexts.
Adverse event profile: No major safety signals at typical investigational doses, but the long-term profile is genuinely not characterized.
Verdict: Investigational. Mechanism plausible, animal-strong, human-sparse. Reasonable to keep watching. Not yet reasonable to recommend.
Common questions
Why don’t you use star ratings?
Stars compress an answer that doesn’t compress well. The difference between a 3-star and 4-star peptide is not the same as the difference between Investigational and Promising. We picked categorical states because they preserve meaning that scales destroy.
Why is “Insufficient evidence” treated as a real verdict?
Because most peptides in public discussion are sitting in this state, and pretending otherwise is the most common way peptide guides mislead readers. Our brand promise is: we will tell you when we don’t know.
Do you accept submitted evidence from peptide brands?
Yes — sent to [email protected] as links to published studies. Brand-prepared white papers, internal data, and unpublished results don’t move verdicts.
Can a verdict be appealed?
Yes. If you believe a verdict is wrong, write to [email protected] with the evidence you think we missed. We read every one. We do not change verdicts because someone is angry, but we have changed verdicts because someone showed us a paper we had not weighted properly.
Last revised: 2026-04-24 — first published version