← All bylines
Desk

Methodology & Standards

Rubric versions, calibration changes, corrections and disclosure notes: the home for "how we score" and "we got this wrong."

Role
Standards desk
Published
6 pieces
Corrections
Contact the desk

Methodology & Standards is the byline attached to rubric-version notes, calibration changes, corrections, and disclosure updates. When a scoring criterion changes, when a band is reclassified, or when an error is corrected, this desk carries the explanation and the evidence behind it. The rest of the masthead is held to the same discipline.

Full scoring methodology

Published under this byline

58 Australian warranty records. Only 3 state labour or service cover without a condition.

The structured register separates product term, labour, freight, registration and transfer. Across 58 exact products, most labour evidence is conditional, excluded or not stated rather than an unconditional service promise.

15 August 2026·6 min read

A specification score cannot know your roof, your grid, or an installer's own record with a product. We have written that limit down.

review.solar's rubric scores exactly what a manufacturer publishes: a specification, a certification, a warranty term, run through the same arithmetic on every product. It was never built to know your roof, your grid connection, or the years a local installer has spent watching one brand outperform another in practice. That limit now has its own line in our methodology, alongside price, bankability tiers and regional certification.

3 August 2026·5 min read

We re-sourced the catalogue against exact-product documents. 70 scores moved, most of them down.

A catalogue-wide evidence pass replaced every editorial estimate with an exact-product source, or withheld the value where none exists. 70 of 122 previously published scores changed, 57 downward. The largest single movement is a pre-release panel whose score was measuring inherited estimates rather than its own record. This note records what moved and why, because a benchmark that corrects itself in public is the product.

1 August 2026·6 min read

An estimate and a datasheet figure used to score the same. They no longer do.

From rubric versions 2.9.0, 3.6.0 and 1.2.0, a criterion recorded as an editorial estimate is pulled 25% of the way toward the non-disclosure floor. 37 of 122 published scores moved, none by more than five points, and no product changed class. This note records what changed and why, because a score you read last week may read differently today.

31 July 2026·5 min read

Three names on a spec sheet. Only one of them is about quality.

BNEF Tier-1, Kiwa PVEL and ITP Renewables get quoted interchangeably in Australian solar marketing as though they all mean the product is good. They measure three different things, one of them is not about the product at all, and one of the programmes has finished. What each is genuinely decisive for.

29 July 2026·8 min read

Battery rubric v2.3: three structural changes and the evidence behind them.

LMFP and sodium-ion split into separate bands. V2G earns a new 90-point tier. HTW Berlin RSP/SPI becomes the preferred efficiency source. Here is the peer-reviewed evidence behind each decision.

1 July 2026·9 min read