OK, real world translation for those of us who aren't analytical chemists: When comparing COAs from different labs, you cannot directly compare purity numbers unless the methods are similar. What constitutes a "good enough" method on a COA? - Detection at 214-220nm (not just 280nm) - Column at least 100mm long - Gradient elution (not isocratic) - Run time at least 15 minutes (shorter runs sacrifice resolution) When a purity number might be misleading: - "99.5% by HPLC" with no method details → unverifiable - Purity at 280nm only → might be missing impurities - Very short run time (<10 min) on HPLC → likely missing peaks - Isocratic method → limited separation capability Bottom line for consumers: A 97% purity result from a thorough UPLC analysis is more trustworthy than a 99.5% result from a quick HPLC screen. Don't chase the highest number — evaluate the quality of the analysis behind it.

Perfect practical summary Pete. Let me add one final point: Lab-to-lab variability is normal. Even with identical methods, different labs using different instruments, different reference standards, and different analysts will get slightly different results. Expect ±1-2% variation in purity measurements between labs. If Lab A says 97.8% and Lab B says 98.5% for the same sample, those results are consistent. If Lab A says 97.8% and Lab B says 92.1%, something is wrong — either with the sample, the method, or the lab. When evaluating a vendor over time, consistency matters more than any single result. A vendor whose products consistently test at 97-99% purity across multiple independent labs is more reliable than one whose results swing from 94% to 99.5% between batches. Thanks to ChromatographerKate and everyone for the expert input. This thread should be stickied — it's reference material for anyone trying to make sense of COA data.

Happy to contribute. One last thought: the analytical testing landscape for compounded peptides is rapidly evolving. The FDA's increased scrutiny of compounding pharmacies is pushing the industry toward more standardized methods. I wouldn't be surprised if we see a USP general chapter or FDA guidance document specifically addressing analytical methods for compounded GLP-1 receptor agonists within the next 12-18 months. That would be a huge step toward making COA comparisons more meaningful. Until then, use the heuristics we've discussed here: check the method, understand the limitations, and don't compare numbers across different methods without context. An informed consumer is a safer consumer.