Version 1.1 took effect September 22, 2026. This transparency revision adds the two-score policy; it is not a silent rewrite. It applies to every Audio review, guide and comparison.
Spec sheets describe capability; only documented listening — with the source device, codec and fit noted — describes experience. An article that cannot say what was actually heard says so, instead of dressing specifications up as impressions.
Bluetooth codecs, ear-tip seal and background noise change what a headphone sounds like more than driver size does. Direct reviews record the full chain; research reviews state which links are missing rather than guessing.
“40-hour battery” means nothing without the volume, codec and ANC state attached. We print the conditions alongside the number, or we do not print the number. The same applies to ANC claims: steady drone, voices and sudden sounds are different tests.
A research-based score ranks relative judgment from consensus and trade-offs — it is not a lab result and never pretends to be one. Readers deserve to know exactly what a number is built from.
Fit and long-session comfort: clamp force, earcup depth, heat buildup and glasses compatibility — the factors that decide whether headphones survive hour three, drawn from owner reports rather than showroom impressions.
Noise control by noise type: steady low-frequency drone behaves differently from voices and sudden sounds; strong ANC on engine rumble can coexist with ordinary conversation handling, and we say which is which.
Codec and connection reality: which Bluetooth codecs are genuinely supported, how multipoint behaves in practice, and which core features depend on the companion app.
Battery claims with conditions: volume level, ANC state and codec attached to every hour rating — without these, battery figures are marketing.
Microphone and call quality: often the weakest link in otherwise strong headphones; assessed from owner reports and published tests, never from the spec sheet.
Price-cycle awareness: previous-generation flagships create value windows once successors ship; we verify the live price and seller rather than assuming the deal still holds.
Hands-on lab-test score: Used only after the editorial team physically tests and measures the product under a documented setup.
Research-based editorial score: Editorial judgment from published review consensus, specification analysis and third-party hands-on reports — not a lab-test result. It follows the article’s relative ranking plus its cited pros, cons, verdict and comparisons; products without an honest differentiating basis are marked Not scored.
Methodology pages rot. PC Gamer’s how-we-test page went 1,583 days without an update — still listing obsolete test rigs — while its real methods lived inside individual guides. That is the failure this page is built to avoid: every methodology here is versioned, dated and reviewed on a schedule.
Evidence honesty, stated plainly: we report the sources reviewed and label manufacturer claims. A methodology should show what supports a conclusion instead of implying product access or measurements that did not occur.
