About Peptide Dosing Protocols

Source Hierarchy

Not all evidence is equal. I grade every data point by where it comes from and label it accordingly:

When you see a dosing number, side effect percentage, or efficacy claim on this site, it comes from Tier 1 or Tier 2 unless explicitly stated otherwise. I do not use supplier marketing pages, anonymous forum posts, or social media influencer claims as evidence sources.

Research Pipeline for Each Protocol Page

Every protocol page starts the same way:

1. Literature search. I search PubMed, ClinicalTrials.gov, and NEJM/Lancet/Nature Medicine for published trial data on the compound. I read the full papers, not just abstracts.
2. Data extraction. I pull specific numbers: dose ranges, titration schedules, half-life values, side effect incidence percentages, and primary efficacy endpoints. Every number gets a source citation.
3. Reconstitution math. I calculate concentrations, dose volumes, and syringe units for every common vial size. Then I cross-check the math at least once before publishing. The PepPal calculator serves as an additional verification tool.
4. Competitive analysis. I review the top 5-8 ranking pages for the same compound to identify content gaps — missing reconstitution tables, absent clinical citations, no comparison data. The goal is to produce the most complete reference available.
5. Drafting with AI assistance. After research is complete, I use AI (Anthropic's Claude) to help draft prose sections from my structured research notes. This is explained in detail below.
6. Human review and fact-check. I read every sentence against primary sources. If a claim doesn't trace back to a citable source, it gets cut. If the math doesn't verify, I redo it. If the tone reads like marketing copy, I rewrite it.
7. Publication and ongoing maintenance. Pages are published and then reviewed whenever new trial data, regulatory changes, or pricing shifts occur. The "Last reviewed" date on each page reflects the most recent substantive review.

Reconstitution Math Methodology

Every reconstitution table on this site is built from three verified inputs:

1. Vial size — confirmed against common grey-market supplier listings and, where available, clinical trial formulation data
2. BAC water volume — selected to produce clean, practical concentrations (typically 1-2 mg/mL for larger peptides, 100-200 mcg/mL for smaller peptides)
3. Resulting concentration — calculated as: Total peptide (mcg) ÷ BAC water volume (mL) = Concentration (mcg/mL)

From the concentration, I calculate dose volumes and syringe units:

• Dose volume (mL) = Target dose (mcg) ÷ Concentration (mcg/mL)
• Syringe units (U-100) = Dose volume (mL) × 100

Every table shows this math so you can check it yourself. The free PepPal calculator automates this for any vial size and water volume combination.

What AI Does

• First-draft generation. After I complete the research phase — reading clinical trials, extracting data points, building reconstitution math — I use Claude to help draft prose sections from my structured research notes. This is significantly faster than writing 4,000+ words from scratch for each protocol page, and it lets me focus more time on data verification and quality control.
• Structural consistency. PDP's protocol pages follow a standardized 14-section format across 24 compounds. AI helps maintain that consistency, ensuring every page covers the same data categories in the same order with the same depth.
• Math cross-checking. I use AI as a second verification layer on reconstitution calculations. If the math doesn't match between my manual calculation and the AI's output, I investigate before publishing.

What AI Does Not Do

• Research decisions. I choose which clinical trials to cite, which data points to include, and how to grade evidence quality. The source hierarchy on this page is my framework — not an AI's.
• Dosing numbers. Every dose range, titration schedule, and cycle length comes from a published clinical trial or a clearly labeled community protocol. AI does not generate dosing data.
• Supplier evaluations. The supplier ratings referenced on this site come from Finnrick Analytics, an independent third-party testing service. AI does not assign quality ratings.
• Editorial judgment. I decide what gets published, what gets cut, what needs a stronger disclaimer, and what evidence tier to assign. Every page is reviewed by me before it goes live.

The Process in Practice

The pipeline for every page: I do the research → I compile structured notes with source citations → AI helps draft from those notes → I review every claim against primary sources, verify all math, and edit for accuracy and tone → I publish.

This is a researcher and software engineer using AI as a writing tool inside a controlled editorial process. The data is mine. The judgment is mine. The accountability is mine.

Google's own guidance says it clearly: content quality matters more than how content is created. AI-assisted content with genuine expertise, rigorous fact-checking, and human editorial oversight is treated the same as fully human-written content. I'd rather show you exactly how the process works than obscure it.