ProblemChild
Explorer
Came across this article while beginning to learn about peptides, and I found it useful. Hopefully it helps other newcomers too.
Comparing COAs Across Different Labs
- Thread starter ProblemChild
- Start date
ProblemChild
Explorer
That's a shame. Perhaps I'm still too inexperienced. I came across it on peptide critic. The article shows up when you hover over the text. Maybe someone with more experience would be nice enough to share it.
SeaGypsy
Explorer
This made for a fascinating read today. Thanks for posting this 
SeaGypsy
Explorer
For me it was fine—simply tap the blue "this article" text at the start of the first line
ProblemChild said:
That's a shame. Perhaps I'm still too inexperienced. I came across it on peptide critic. The article shows up when you hover over the text. Maybe someone with more experience would be nice enough to share it.
Nope, you got it right, I just didn't catch the blue linked text.... my vision has declined since I began glp-1s, though that's merely a cop-out....
It would've made the article more intriguing had Krause (Finndicks' go-to lab) been part of it. They'd likely have claimed the vial contained no reta, or 3mg.
BNLFL
Explorer
The blue one also slipped past me.
SeaGypsy
Explorer
Fingers crossed mine improve too — can barely see
ProblemChild
Explorer
I put the text in bold. My hope is that this makes it easier.
CNCCurrency
Explorer
Newsletters go out to subscribers on a regular basis from him — or more likely, it's Randy the rat who does it.
BNLFL
Explorer
Randy seems fine to me, though he's not exactly animated or talkative.
Flash-BCR
Explorer
That was a really engaging piece of content... 
eidos
Explorer
A couple of their arguments do hold up: calibration matters, and fill volumes can differ even inside one kit. Where they go wrong is in judging what an honest reading of the data actually requires — they even buy into the illusion of precision those numbers project.
Let me lay out the reasoning (nothing beyond high-school math, applied to biochemistry).
Three things matter most:
1. Instrument bias: this is the reason calibration exists in the first place.
2. Uncertainty: any analytical setup shows “natural” scatter in its output. The usual practice is to report a ± figure (not one lab does this).
3. Reproducibility: you run the same source repeatedly, then compute a mean and a standard deviation (sigma). No lab brings this up.
On top of that, outcomes shift with the instrument maker, the generation of the device, which solvent is used for elution, the column type, and how concentrated the peptide is: the weaker the concentration, the more peptide — lipophilic chain and all — stays stuck to the tubing and the column (hence washing or replacing columns between runs).
Go to a reference supplier like Merck-Sigma and you’ll find aGLP-1 peptides offered at 95% or 97% purity. Check what they charge, if you want a good laugh.
The COAs floating around today ought to honestly say 99% purity rather than 99.834 or 99.971 — that’s pure sleight of hand. HPLC-UV precision lands somewhere between 0.1 and 0.5, depending on settings and how many runs you do.
Peter Magic at least admitted he reports results that way for marketing. He’s running a business, 40 employees, equipment to pay off. With 1,000 tests a day, every day of the year, I’d be shocked if they did more than a single run unless a customer requested and paid for more.
From what I know about production management in pharma, and from the information out there, I’d guess China has under ten supplier factories. Batch numbers never show up — only SKUs — so tracking stays hard. What looks increasingly clear is that each peptide gets made in just 1 or 2 batches per year. That’s why sellers hand out COAs dated 2025. Consider how many kits 10 or 20 kg can yield.
Most likely the same batch gets tested dozens or even hundreds of times. You can’t tell because of the fake precision — 4 or 5 digits where 2 significant figures belong. Testing labs keep quiet about this: it’s marketing.
At 99%, manufacturers are meeting good manufacturing practices. Their product is fine. Sterility or endotoxin problems were rare across all batches. Such problems show up in the official drug supply too, with recalls happening daily.
The one genuine issue is commercial, not a production matter: getting scammed by someone who ships mannitol with no peptide. Luckily, buyer reviews here help you dodge that.
Let me lay out the reasoning (nothing beyond high-school math, applied to biochemistry).
Three things matter most:
1. Instrument bias: this is the reason calibration exists in the first place.
2. Uncertainty: any analytical setup shows “natural” scatter in its output. The usual practice is to report a ± figure (not one lab does this).
3. Reproducibility: you run the same source repeatedly, then compute a mean and a standard deviation (sigma). No lab brings this up.
On top of that, outcomes shift with the instrument maker, the generation of the device, which solvent is used for elution, the column type, and how concentrated the peptide is: the weaker the concentration, the more peptide — lipophilic chain and all — stays stuck to the tubing and the column (hence washing or replacing columns between runs).
Go to a reference supplier like Merck-Sigma and you’ll find aGLP-1 peptides offered at 95% or 97% purity. Check what they charge, if you want a good laugh.
The COAs floating around today ought to honestly say 99% purity rather than 99.834 or 99.971 — that’s pure sleight of hand. HPLC-UV precision lands somewhere between 0.1 and 0.5, depending on settings and how many runs you do.
Peter Magic at least admitted he reports results that way for marketing. He’s running a business, 40 employees, equipment to pay off. With 1,000 tests a day, every day of the year, I’d be shocked if they did more than a single run unless a customer requested and paid for more.
From what I know about production management in pharma, and from the information out there, I’d guess China has under ten supplier factories. Batch numbers never show up — only SKUs — so tracking stays hard. What looks increasingly clear is that each peptide gets made in just 1 or 2 batches per year. That’s why sellers hand out COAs dated 2025. Consider how many kits 10 or 20 kg can yield.
Most likely the same batch gets tested dozens or even hundreds of times. You can’t tell because of the fake precision — 4 or 5 digits where 2 significant figures belong. Testing labs keep quiet about this: it’s marketing.
At 99%, manufacturers are meeting good manufacturing practices. Their product is fine. Sterility or endotoxin problems were rare across all batches. Such problems show up in the official drug supply too, with recalls happening daily.
The one genuine issue is commercial, not a production matter: getting scammed by someone who ships mannitol with no peptide. Luckily, buyer reviews here help you dodge that.
ProblemChild
Explorer
A Reddit link here provides a summary of a Janoshik YT interview.
TLDR:
Janoshik got its start largely by chance, yet grew into one of the primary labs for peptide testing. According to Peter, peptides now account for 70–80% of their business, roughly 100 peptide samples a day are processed, and the key point is: blind/community testing matters way more than trusting vendor COAs.
TLDR:
Janoshik got its start largely by chance, yet grew into one of the primary labs for peptide testing. According to Peter, peptides now account for 70–80% of their business, roughly 100 peptide samples a day are processed, and the key point is: blind/community testing matters way more than trusting vendor COAs.