1. Who tested it?
The lab is named at the top of the report. Is it an independent lab you can find online, with its own website and address? Or a report from the manufacturer itself? A manufacturer report is better than nothing. It is not independent.
2. Does the report match your vial?
Compare substance, stated amount and batch number on the vial with the report. Check the date too. A report from two years ago says little about a batch filled last month.
3. Is the report genuine?
Many labs let you verify their reports. Janoshik, for example, prints a task number and a verification key on every report. With both you can check on janoshik.com/verification whether the report really exists there.
If there is no such code, ask the shop for the original PDF or write to the lab directly. A reliable shop has no problem with that.
4. Reading the numbers
Purity (HPLC) shows how much of the sample is the substance in question. Values of 98 to 99 % are common.
Content shows how many milligrams are really in the vial. If the label says 10 mg and the lab measures 9.6 mg, that is 4 % less. That is fine. At 7 mg, 30 % is missing.
Identity (mass spectrometry) shows whether it is the right substance at all. A sample can be 99 % pure and still contain only 6 mg. That is why you need both values.
5. Warning signs
No batch number. The same report for every size. A blurry photo instead of a PDF. Numbers like 100.00 %. A lab name that doesn't exist online. Fonts that change mid-document. The more of these come together, the less we would trust the report.
What a COA does not show
A standard report says nothing about sterility, endotoxins or heavy metals. Those need separate tests. And it only applies to the sample that was measured. A good report is a good sign. It is not a guarantee.