Nobody paid for these scores
Every product is scored against tables we publish — so you can see what each number means, and where the evidence ran out.
- Anchor tables published
- Paid placement never changes a score
- Depth in one category before the next
Usability
Our own assessment, against eight criteria. Each one is scored from a fixed table, so a figure means the same thing on every product we cover. It reads specifications, certifications and documentation.
Does not read reviews.
Social
What independent sources actually say, across the five places people already look: Reddit, web coverage, video, Amazon and retailer reviews. Sources we cannot verify are excluded.
Does not read specifications.
The numbers we would rather not publish.
A score is only worth anything if the gaps are visible too. Nothing here is entered to fill a column, and a product we cannot assess says so on its own page.
The criteria do not describe them, or the manufacturer publishes too little to assess. Both render as Not scored, with the reason stated.
Two populated sources is the floor. One is an anecdote wearing an average, so those products get no social score rather than a flattering one.
Where a manufacturer does not publish a specification, that criterion scores nothing. We do not estimate it and we do not quietly drop the row.
What doesn't count
These all look like evidence. They aren't, for our social score.
- Brand's own widget
- Syndicated reviews
- Brand-store aggregates
- Five-star-only counts
- Seller ratings
- Rivals reviewing rivals
Want the full method?
Weights, floors, blanks, corrections, and source rules live on How We Score. The cards above are the live shape of a score; that page is the rulebook.
