Most supplement labels tell you what is supposed to be in the tub. A Certificate of Analysis is supposed to tell you what a laboratory actually found. The gap between those two things is where a great deal of money in this industry gets made, and reading a COA properly is a skill worth about ten minutes of your time.
Here is what to look at, in the order that matters.
Start with the client name, not the purity figure
A real report names who commissioned the test. If that field is blank, redacted, or belongs to a company with no relationship to the seller in front of you, the report may be perfectly genuine and still have nothing to do with the product in your cart.
This is the most common trick in the category: a real COA for somebody else’s batch.
Then the batch or lot code
The report should carry a batch code, and that code should appear on the product you receive. A COA with no batch reference is a document about nothing in particular — it cannot be matched to anything, which is often precisely why it was published.
Then ask where the report lives
This is the one that separates a checkable seller from an unchecked one. A PDF hosted on the seller’s own site proves little, because nothing prevents it being edited before upload. What counts is a report that resolves on the testing laboratory’s own portal, where you enter the code and the lab’s system returns the result.
If you take one thing from this article, take that. It is the difference between trusting a laboratory and trusting a seller’s copy of a laboratory’s work.
Then the test date
A COA is a snapshot of one batch at one moment. A report from two years ago describes a batch that is long gone. Look for a date, and look for whether the seller publishes new reports as new batches arrive or has simply left the same handful of files up since launch.
The purity figure comes last
By the time you reach the number you already know whether it means anything. And when you do read it, read the whole set rather than the headline. A supplier publishing only flawless results is either exceptional or selective, and a range that includes some unflattering figures is the more believable record.
A Canadian worked example: Growth Guys publishes 236 reports through Janoshik Analytical across 132 products, with 175 stating a purity figure. The average is 99.03%, the range runs 93.84% to 99.99%, and 32 results sit below 98%. Those below-average results are the reason the set reads as a record rather than marketing.
What a COA does not cover
Worth being straight about this. A purity figure describes what was in a sample the seller sent to a laboratory. It says nothing about what arrives at your door — that is a separate question only independent blind testing answers, and almost nobody in this category commissions it. It also says nothing about whether a compound is appropriate for you, which is a conversation for a qualified professional rather than a document.
A habit worth building
Whatever you are buying, test one public claim you can check before trusting the ones you cannot. With supplements the easiest is usually a discount code: open the store signed out, add an item, enter the code at the cart stage before payment, and read the discount line against the subtotal.
Growth Guys publishes HEALTHYLIFE10 for 10% off sitewide with no minimum purchase and no product exclusions, and unusually, publishes the dated readings behind it at growthguyshealthylife10.ca/archive.html rather than simply asserting that it works.
A seller who publishes checkable evidence for the small claims is more likely to be careful with the large ones. That correlation is not proof, but it is close to the best signal available from outside.
For a scored comparison of Canadian suppliers against a published rubric, canadianpeptidereviews.ca applies the same criteria to each and shows its working.
Client name, batch code, where the report lives, the date, then the number. In that order, every time.
