kate.chem said:

The provenance questions are more useful than the numbers.

COA red flags for certificates of analysis — here's what to look for when reviewing a Certificate of Analysis:

• No HPLC data: If there's no chromatography graph, it's not a real COA
• Exact 100% purity: Suspicious — real testing always shows slight variation
• No batch number: Can't trace back to specific production
• No endotoxin testing: For injectables, this is essential
• Template looks like it was made in Word: Professional labs have standardized formats

I have seen fake COAs that are literally photoshopped. Always verify independently if possible.

A narrower follow-up, since the general answer is now clear:

Whether a purity figure and a content figure can disagree without one of them being wrong?

dave_SLC said:

COA red flags for certificates of analysis — here's what to look for when reviewing a Certificate of Analysis: No HPLC data: If there's no…

HPLC results interpretation for certificates of analysis: when you get your Janoshik report back, here's what the numbers mean:

• Purity %: Should be >97%. Below 95% is concerning.
• Retention time: Should match reference standard for the peptide
• Related substances: Degradation products should be <2% total
• Single impurity: No individual impurity should exceed 1%

My last batch: 99.7% purity, 0.3% total related substances. Clean chromatogram with single sharp peak at expected retention time. Exactly what you want to see.

Closing the loop on my own question.

Purity and content being different measurements was the piece I was missing. The vial was overfilled and pure, which is a very different problem from the one I thought I had.

PharmHunterJen said:

HPLC results interpretation for certificates of analysis: when you get your Janoshik report back, here's what the numbers mean: Purity %: Should be…

Agreed, and mass spec answers a question purity does not: identity. High purity tells you one thing dominates the chromatogram, not that it is the molecule you ordered. A mass within about a dalton of expected is the identity check.