How MATCHLAB works.
Decisions first, products second. Here is exactly what we do and what we refuse to do.
1. We start from a decision, not a keyword
Every guide answers one real buying question, such as which kind of vacuum suits a home with pets and rugs. We do not publish a page for every phrase someone might type. One product type gets one guide, and a shortlist page only where real listings add something.
2. What a guide is built from
A selection guide covers a product type, not individual models: how the main designs differ, which specifications change the experience, and what to check on a listing. It is written from general knowledge of the category, not from reading specific products' manuals or customer reviews, and it says so on the page. Shortlist pages are different: every fact about a product there comes from that product's own Amazon listing, and the names and images are rechecked against the live listings.
3. How guides are written and checked
Guides are drafted with writing software against a fixed brief, then pass an automated quality gate before they go live: minimum depth, no testing claims, no prices or ratings, no attributed statistics, no duplicated wording across pages, and a check that the metadata is unique. Published guides are read by an editor afterwards and corrected or withdrawn when they fall short. We publish a small number of guides per day on purpose.
4. What we will never do
We do not invent test results, star ratings, prices or user reviews. We do not claim to have tested something we have not tested. We do not let commissions decide the order of recommendations.
5. How we earn
Some links go to Amazon, and as an Amazon Associate we earn from qualifying purchases. The price you pay is the same. Read the full affiliate disclosure.
About the product finder
The finder on the homepage shows example product types to help you think about tradeoffs. It does not show live listings, prices or ratings, and it does not use a trained recommendation model.