About | Verdict

What this site is

Verdict publishes three things. A review covers one product and ends in a number. A comparison puts two or more products through the same tests and picks one. A guide is everything else — the roundups, the buying advice, and the explanations that stop the other two having to re-explain what a spec sheet means.

How we make money, and what it does not buy

We earn a commission when a reader buys through a retailer link. That is the whole business model, and it has two consequences we are explicit about.

First, a commission is paid on any product a reader buys through us, not on the one we recommend — so there is no financial reason to push an expensive pick over a cheap one. Second, testers do not see the affiliate figures, and the price of a unit is withheld from the person testing it until the scoring is finished. Neither of those is a guarantee of virtue. They are just the arrangements that make the obvious failure modes harder.

We buy what we test. When a manufacturer sends a loan unit — which happens, usually for a product that has not gone on sale yet — the review says so on its face, in the same line that says how long we had it.

What the number means

Every review publishes a weighted breakdown, and the headline score is the weighted average of it. The theme computes that arithmetic rather than trusting a typed-in figure, so the badge can never disagree with the table under it. Scores are on a 0–10 scale and land in one of five bands:

The bands

9.0–10.0ExceptionalBest in its class. We buy this with our own money.
8.0–9.0Strong buyRecommended without caveats for most people.
6.5–8.0Good, with caveatsWorth buying if the compromises below suit you.
5.0–6.5Only on saleHard to recommend at list price.
0.0–5.0Skip itSomething else does this better for less.

Corrections and re-tests

A recommendation is only as good as its last check. Guides carry a revision date and a note saying what changed; reviews get re-opened when firmware, pricing, or a rival moves enough to change the answer. If we quietly stop recommending something, the note says that too — a withdrawn pick is information, not an embarrassment to be hidden.

Found something wrong? Tell us. Corrections are made on the page, dated, and described.

Who does the testing

All reviewers

Priya RaghunathanSenior reviewer, audio8
Runs the audio bench: a fixed head-and-torso rig, a sealed room, and a commute she has taken so many times it counts as a control group.
Dana WhitlockEditor-in-chief7
Editor-in-chief. Sets the test protocols, signs off every score, and keeps the spreadsheet where a product's price stays hidden until the testing is done.
Marcus OyelaranLead reviewer, computing6
Nine years of shipping laptops back to manufacturers with the battery report attached. Cares more about sustained clocks than peak ones.
Nils HalvorsenReviewer, fitness and wearables5
Puts roughly 60 miles a week into whatever the lab sends him, with a chest strap on and a track that does not lie about distance.
Sofia AlmeidaTest lead, kitchen5
Pulled espresso professionally for six years before writing about the machines. Owns four thermometers and trusts two of them.
June ParkReviewer, workspace and ergonomics4
Assembles every chair and desk twice — once by the manual, once the way a tired person would at 9pm — and reports both experiences.