Most product reviews are easy to evaluate once you know what to look for. These six checks take about a minute each and will tell you more than reading the review twice.
1. Who wrote it, and who pays them?
Every review has a commercial relationship behind it. An affiliate site earns when you click through. A retailer earns when you buy. A brand-owned site — like this one — earns when you buy from the brand.
None of those is automatically disqualifying. What matters is whether the relationship is stated plainly and near the top, or buried. A site that describes itself as independent while earning commission on every link is telling you something about its standards before you read a word of the review.
2. Does it separate what was tested from what was read?
A review based on the product page and a review based on eight weeks of use are different documents, and a good review says which it is.
The tell is sensory detail. Texture, scent, absorption, how it layers, how the pump behaves — these can only come from handling the product. If a review describes all of that but never says when or how it was tested, be sceptical. Formula review versus hands-on review covers the distinction in full.
3. Does it say what it does not know?
This is the single most useful check. Every cosmetic product has attributes the brand does not publish — often the concentration of the marketed active, often the full ingredient list.
A review that presents a complete-looking picture of a product whose maker publishes an incomplete one has filled the gaps with something. Usually that something is a plausible guess.
4. Are prices and formulas dated?
Prices change weekly. Formulas change without announcement. A review quoting a price with no date attached is quoting a price that may be two years old.
Look for an explicit "last checked" date, and look for whether it is recent. An undated review is not necessarily wrong; it is unverifiable, which for a factual claim amounts to the same thing.
5. Is ingredient research being passed off as product evidence?
This is the most common and most convincing failure mode. A review cites a real study about an isolated ingredient, then discusses a product containing that ingredient as though the study were about the product.
The study controlled concentration, delivery and duration. The product controls none of them in the same way, and usually does not publish its concentration at all. The citation is real and the inference is not. Watch for a paragraph that opens with research and closes with a product recommendation, without acknowledging the step in between.
6. Is there anything real in the limitations?
Almost every review has a limitations section now. Most of them are empty in substance: "the packaging could be more premium", "we wish it came in a larger size".
A real limitations section names something that would genuinely stop some readers buying — an undisclosed concentration, a missing ingredient list, a claim the brand cannot support, an inconsistency between the brand's own pages. If nothing in the limitations would change anyone's decision, the section is decoration.
Applying this to a review on this site
Try it on our GHK-Cu + SNAP-8 Peptide Renewal Serum review. Ownership is disclosed at the top. The review type badge says it is a specification review, not a test. Every price carries a check date. The unknowns section lists five things we could not establish. If any of that were missing, the checks above should have caught it.