The difference between a useful review and a frustrating one is rarely expertise. It is specificity: whether the author can tell exactly what to change, where, and why.
Anchor every finding to the text
Quote the passage. A finding attached to specific wording is unambiguous; the same point made in general terms sends the author hunting through their own manuscript trying to guess what you meant.
Separate what must change from what could
Marking a finding major, minor or a note is not decoration. It tells the author where to spend limited revision time, and it stops a stylistic preference from reading with the same weight as a flaw in the analysis.
Reserve major for things that would draw a referee’s objection. If everything is major, nothing is.
| Severity | Use it when | The author should |
|---|---|---|
| Major | A referee would object to this | Fix it before submitting |
| Minor | It weakens the paper but would not sink it | Improve it if there is time |
| Note | A suggestion, not a defect | Take it or leave it |
Say why, not just what
A change the author does not understand is a change they will implement badly or argue with. One sentence of reasoning turns an instruction into something they can apply to the rest of the paper themselves.
Write to someone who will read it
Anonymous refereeing has a reputation for brusqueness that helps nobody. This review goes to a named person who paid for it and can rate it. Plain, direct and specific is what works; contempt is not rigour.
Common questions
How long should a review be?
Long enough to cover what matters. Five precise findings beat twenty vague ones, and an author is better served by a short review they can act on than a long one they cannot.
Should I recommend accept or reject?
Neither: this is not an editorial decision. You give the author your honest read on whether the paper is ready, needs minor revision, needs major revision, or needs rethinking, and that goes to the author only.