A good review does two jobs at once. It tells you what the writer heard, and it tells you why that should matter to you. Most bad reviews fail at the second job. They hand you an opinion with no evidence under it, and you are left agreeing or disagreeing for no better reason than mood.
The trick to reading like a critic is separation. Pull the taste apart from the substance. Taste is the verdict: this record moved me, this concert dragged. Substance is the case: which track, which bar, which performance choice produced that reaction. A review can have strong taste and weak substance, and once you see that split clearly, you can use even a bad review well.
This matters whether you are deciding what to stream tonight or deciding whether a writer's judgment is worth following for a year. The same skill applies to album write-ups, concert notices, and even gear coverage — our own reviews section works on the same principle, evidence first, verdict second.
What is the difference between taste and substance?
Taste is preference. It is the part of a review that cannot be argued with, only shared or not shared. "The singer's tone leaves me cold" is taste. It is honest, it is useful, and it is not wrong even if you love that singer.
Substance is everything the writer offers in support: what was played, how it was played, and what a listener with different tastes could still check for themselves. "The singer flattened the top of the phrase in the second verse, and the band's tempo sagged to meet it" is substance. You can hear it or not hear it, but you can locate it.
The healthy ratio is roughly one verdict to several verifiable observations. When a review runs the other way — five adjectives for every concrete detail — you are reading taste dressed up as judgment.
What should you look for first in any review?
Look for specifics you could point to on a clock. A timestamp. A track number. A bar. "The bridge at 2:40" gives you something to test. "The whole album feels long" gives you nothing.
Merriam-Webster lists one meaning of "read" as "to interpret (a musical work) in performance" — and that is exactly what a critic is asking you to do in reverse. The performer interprets the score; the reviewer interprets the performance; you interpret the reviewer. Each layer should be checkable against the one below it. If a writer says a pianist rushed the finale, you can go listen to the finale. If a writer says the pianist "fails to connect emotionally," there is nothing to check. That is not a crime — some fine critics write that way — but it is taste, and you should file it as such.
Practical steps: on your first pass, underline every claim you could verify by listening. Then underline every claim that is pure preference. If the second pile is much bigger than the first, you have learned something about the writer, not the record.
How do you spot the signals of a well-made review?
Well-made reviews share a few habits, and once you know them you can spot them in a paragraph.
- Named details. The soloist, the engineer, the venue's acoustic, the tempo choice. Vague praise — "stunning," "breathtaking" — is a warning sign in any review, including ours.
- Compared choices. "She takes the second movement slower than the standard reading, and the slow tempo exposes the inner voices." A comparison is evidence, even when the comparison is only to the score or to the performer's own earlier recording.
- Admitted limits. A critic who says "the live recording does the hall no favors" is telling you the boundaries of the evidence. That honesty is a trust signal.
- Costs named. Every interpretive choice trades something. A faster tempo buys momentum and costs clarity. A review that names the trade-off is doing the work; one that names only the win is selling.
Our analysis across years of reading criticism in the classical and guitar press: the single best predictor of a review's usefulness is whether it names where an interpretation changes. Not what the writer felt in the abstract, but the exact moment the reading turned. That is the detail a generic listener would not reach, and it is the detail you can go hear. This connects to our earlier piece, How to Choose a Guitar Humidifier: Specs, Prices, Trade-Offs.
How should you weigh a reviewer's taste against your own?
Calibrate, don't copy. If you read a writer long enough, you learn their leanings. Maybe they favor dry, close-miked recordings over spacious hall sound. Maybe they reward risk-taking even when it fails. None of that is a flaw. It is data.
Once you know a critic's taste, you can invert their verdicts usefully. "This writer loves sparse arrangements and hated this record — it is probably lush, and I like lush." That is not gaming the system. It is the same skill record shoppers have used forever: find the reviewer whose dislikes map to your likes.
What you should never do is treat a review as a substitute for listening. Reviews are maps, not territory. A map drawn by someone with different feet than yours can still be an excellent map. It just will not tell you which route you will enjoy.
What about reviews of gear and instruments?
The same split applies, with one difference: gear has measurable substance. Scale length, nut width, pickup type, street price. A gear review earns trust by putting those numbers on the table with their source, then keeping the sound descriptions clearly labeled as impressions. When a review says "the maker describes this pickup as warm," that is a sourced claim about the maker. When it says "this pickup sounds warm to me," that is taste. Both can appear in the same honest review — the difference is the label.
Watch for the tell of a review that never separates the two. If every sonic claim arrives with the same confidence and none carries a source or a first-person label, the writer is either hiding taste inside fact or has not thought about the difference. Neither is a good sign. You can see the labeled approach in practice in something like our Capos Compared: A Sourced Guide, where the measurable facts and the practical judgments sit in separate, clearly marked layers.
Price and availability claims deserve special suspicion, in music and everywhere else. A review that names its price source and date is doing the minimum. One that prints a price with no date or source is asking you to trust memory — usually the writer's, sometimes nobody's.
Practical steps: a three-pass way to read any review
Here is a process, adapted from how working critics actually read each other. It takes about ten minutes.
- First pass: the verdict. Find the recommendation and the reason attached to it. One sentence. If you cannot find a reason, the review has told you its taste and nothing else.
- Second pass: the evidence. List every checkable claim — timestamps, named players, tempo choices, recorded sources, prices with dates. Ask: could I confirm this? Would it change my listening if I did?
- Third pass: the writer. Ask what taste is doing the steering. What does this critic reward? What do they punish? Note it. Over a few months, your notes become a personal map of whose judgment travels well with yours.
The third pass is the one most readers skip, and it is the one that pays off over time. A single review tells you about an album. A pattern across twenty reviews tells you about a critic, and that is the more durable asset.
Where this leaves you
None of this turns you into a critic, and it should not. It turns you into a reader who knows which parts of a review are arguments and which are invitations. The arguments you can check. The invitations you can accept or decline on your own taste, now with better information about whose taste is doing the inviting.
What the evidence above does not settle is the oldest question in criticism: whether taste can be trained. My own reading of the working critics I trust is that taste sharpens the way ears do — through comparison, not assertion. Listen more, compare more, and the verdicts you disagree with become more interesting, not less. That is the real payoff of reading like a critic: not better agreement, better disagreement.
