About AMZ Reviews Data
Who runs this site, how the review-data pipeline works, and what we deliberately do not publish.
AMZ Reviews Data is an independent review-data research publication. Cookware is our first beat: instead of a sponsored “top 10”, we answer specific questions β which pan for induction, what actually fails on a $25 nonstick pan, whether a brand’s durability holds up after two years β using aggregated Amazon review data published transparently.
What makes it different
- One data source, one method. Every score comes from the same offline pipeline and is published on the scoring standards page. Nothing is scored in the browser and no score is hand-written.
- Findings, not descriptions. Anybody can say “aluminum dominates cookware”. We publish the comparisons only our corpus supports, with the metric, the sample size and the limitations attached β see Findings .
- No sponsored rankings. We do not accept payment for placement. Some outbound links are Amazon affiliate links, which pay us a commission at no cost to you; they never change a score or a ranking. See our editorial policy .
- No raw reviews. We publish aggregates, counts, rates and short representative quotes only.
Where the numbers come from
The corpus is 18,106,825 review records covering 1,186,539 products, analysed offline into a single database. 43,093 products have a six-dimension score, of which 21,482 meet the minimum sample threshold; 669 brands have enough evidence to be scored.
The full methodology and its limits are on the Data Sources & Limitations page.
Who runs it
AMZ Reviews Data is an independent research project maintained by a small editorial team. Data is computed offline in Python; every published note is written and reviewed by an editor through the process described in the editorial policy . You can see who reviewed a page in the “Reviewed by” line at the bottom of reviewed pages.
What we deliberately do not do
- We do not buy or physically test products. Our evidence is review-derived, and we say so on every page β see How we work .
- We do not pretend the snapshot is real-time; the data vintage is 2023 and we say so everywhere.
- We do not generate pages just to cover a keyword β every page must pass an evidence and specificity gate before it can ship.
- We do not publish a page per ASIN, brand, store or manufacturer. Cookware is our first beat; other categories get a data note only when the sample supports one.
Questions or corrections? Contact us . See also the editors behind the data.