An album review score at Strum Magazine answers one question: how well does this record do what it sets out to do? It is not a sales rank, a popularity contest, or a verdict on the artist. It is our judgement of the playing, the writing, and the recording, argued in the body of the review so you can disagree with the number and still use the reasoning.
The score is the last thing we write, never the first. It sits at the end of a process that starts with plain listening and ends with a rating on a five-point scale. If the argument in the review does not support the number, the number changes. The review is the product; the score is a convenience for readers deciding what to hear. We covered a connected angle in How to Choose a Guitar Humidifier: Specs, Prices, Trade-Offs.
This page explains the criteria, the scale, and the things we refuse to let into a score. If you want the reader's side of the bargain — how to weigh a critic's opinion against your own ears — our guide on how to read a music review like a critic covers that in detail. Everything below is the writer's side.
What counts as an album in the first place?
The word has drifted for a century. According to Wikipedia's entry on the album, the term began as bound books of 78 rpm records in the early twentieth century, evolved into the vinyl LP after 1948, and now covers CDs, downloads, and streaming releases alike. Merriam-Webster defines it simply as "a collection of one or more audio recordings that is presented as a single unit and is typically longer than a single or an EP."
That definition matters for scoring because we judge the unit, not just the songs. A record is a sequence: running order, pacing, the arc from track one to the closer. A collection of strong songs in a careless order scores lower than a slightly weaker set that holds together. Classical and acoustic releases — the core of our coverage — get the same treatment, with liner notes and documentation weighed as part of the presentation, since those materials carry credits, texts, and translations.
What the score measures
We score four things, in this order of weight.
- Performance. Is the playing and singing committed, controlled, and idiomatic? For classical records this means intonation, ensemble coordination, and phrasing you can follow bar by bar. For singer-songwriter and band records it means timing, dynamics, and whether the performance sounds chosen rather than assembled.
- Material. The writing itself: melody, structure, harmony, and for classical repertoire, the choice and shaping of the programme. A great performance of thin material can only go so far, and we say so plainly.
- Recording and production. Does the sound serve the music? A dry, close studio sound suits some repertoire; a reverberant hall suits others. We judge fit, not gloss. A live album, for instance, is judged partly on whether the room — the audience, the ambience — earns its place on the record.
- Coherence. Does the whole hold together as an album, in the sequence sense above? This is the criterion most often missing from track-by-track write-ups, and it is where reissues and compilations get judged hardest.
Notice what is absent: no criterion asks whether the record is important, fashionable, or likely to chart. Those are facts about the world around the record, not the record.
What the score deliberately ignores
Commercial performance never touches the number. The industry itself separates the two: the Billboard 200, the week's most popular albums as compiled by Luminate, ranks records by multi-metric consumption blending sales, track equivalents, and streaming equivalents — a measure of reach, not of merit. A record can top that chart and earn a middling review here, or sell little and earn our highest rating. Both outcomes are normal.
We also ignore the press narrative around a release. An artist's backstory, a label's marketing push, an anniversary, a comeback — none of it moves the score. It may move the context, because context is part of honest criticism, but the rating answers only to what is in the grooves, or in the files.
Finally, we do not score against the artist's own past as a fixed ceiling. A debut is judged as a debut. A late-career record is judged on its own terms. Expecting record seven to be record two is a critic's error, not a finding.
How the five-point scale works
We use a five-point scale with half steps, and we treat the middle of the scale as a real grade, not an insult.
- 5 — A record that sets a standard in its repertoire. Rare, and the review has to earn the word.
- 4.5 — Excellent, with a specific, named reservation.
- 4 — Strong and recommended, with trade-offs stated.
- 3.5 — Good, worth your time if the repertoire or artist interests you.
- 3 — Solid but uneven; the review names the uneven parts.
- 2.5 and below — The problems outweigh the pleasures, and the review explains exactly where they sit.
Two commitments keep this honest. First, every score above 4 must name its reservation — no perfect record gets a free pass, and no 4.5 hides a flaw the writer disliked. Second, we do not inflate for kindness. A 3 from us means the record works; it is not a failure. Readers who treat anything under 4 as a pan are reading a scale we do not use.
How a review is actually written
Our working method is deliberate repetition. Each reviewer listens to a record at least three times before drafting: once for the sound and the room, once for the phrasing and structure, once for what the performer avoided — the risks not taken, the easy effects declined. Notes come from those passes, not from memory. Only then does the draft begin, and the score is assigned last, after the argument is on the page.
We name what we hear precisely: the exact track where an interpretation turns, the chorus where the arrangement thickens, the bar where the ensemble goes flat. Precision is the reader's tool. "Beautiful" and "timeless" tell you nothing; a timestamped observation tells you where to put the needle.
Every review also states what the listening could not establish. If a pressing sounds inconsistent across copies, we say we heard one copy. If a classical programme pairs a familiar work with an obscure one, we say which judgment rests on which. Honest gaps beat confident bluster, and they are the reason our scores can be argued with rather than merely accepted.
Practical steps: using our scores well
Here is how to get the most out of any review score, ours or anyone's.
- Read the review before you read the number. The reasoning is the reviewable part.
- Check which criteria the critic leaned on. If you care most about production and the critic weighted material, adjust accordingly.
- Treat the middle of the scale as the norm. A 3.5 is a usable record, not a warning.
- Track a critic over time. One score means little; a pattern of agreement or disagreement with a writer means a lot.
- Use scores to build a listening queue, not to close a case. The record itself is the only verdict that counts.
Where our reviews sit in the magazine
Album reviews are one branch of our Reviews coverage, which also spans instruments, amps, and pedals — gear guides built from manufacturer documentation and published measurements rather than impressions. Records get the same discipline from the other direction: no gear to measure, so the evidence is the listening itself, documented in detail. Our Artists and Classical Music sections carry the profiles and repertoire context that a review can only gesture at.
What this methodology cannot settle is taste, and it does not try. It can tell you what a record attempts, how well it succeeds on the evidence of repeated, attentive listening, and where we might be wrong. The rest is your evening, your headphones, and your call.
