A friend hands you a novel and says, "You'll love it." You believe her because you know what she reads, what she dislikes, and how often she is right about you. An online review offers none of that context by default. It arrives as a block of text from a stranger, sometimes with a star rating attached, sometimes with a purchase badge, sometimes with nothing at all. The trick is not to distrust reviews, but to read them the way you would read any other piece of writing: asking what it actually tells you, and what it quietly leaves out.
Most readers learn this by trial and error, usually after buying three disappointing books in a row on the strength of the word "unputdownable." A little method saves a lot of that. Here is what separates a review worth your time from one that is merely loud.
What a Review Is Actually For
A good review does two jobs at once. It describes the book honestly enough that you can decide whether it is for you, and it makes a judgement you can examine rather than simply accept. That second part matters more than readers often admit. Anyone can summarise a plot, and a summary is not a review. The value comes from the reasoning: this book tries to do a particular thing, here is how well it does it, and here is who might care.
So the first question to ask is not "did this person like it?" but "did this person explain why?" A negative review with clear reasons is far more useful than a five-star rave that only reports the reviewer's emotions. If you finish reading and still cannot say what the book does well or badly, you have read a mood, not a review.
Evidence Beats Enthusiasm
Specificity is the clearest signal of a trustworthy review. Vague praise can be written about almost any book; particular observations cannot. Compare these two sentences:
- "The characters are so relatable and the story keeps you guessing."
- "The middle third slows down while the narrator works through two family secrets, and the tension drops because neither secret affects anything until the final chapter."
The second one tells you something you can weigh against your own taste. It may even be wrong, but it is checkable — and that is the point. Look for references to structure, pacing, narrative voice, translation quality in a translated work, or the way a book handles a subject. Look for a quoted line or a described scene. None of this guarantees the reviewer is right, but it shows they read attentively and are willing to be pinned down.
Be suspicious of reviews built entirely from adjectives. "Stunning," "heartbreaking," "powerful" — these words cost nothing. A reviewer who leans on them for five paragraphs is often describing the marketing copy rather than the book.
Who Is the Reviewer Writing For?
Fit is everything in book reviewing, and good reviewers tell you where they stand. A thriller fan reviewing a slow literary novel is not useless, but their complaints about pace need to be read in context. When a review opens with something like "I read almost exclusively historical fiction, so I came to this expecting…", you can adjust accordingly.
Ask yourself three things:
- What does this reviewer normally read? A short profile, a few of their other reviews, or a line in the introduction usually answers this.
- Do they describe the intended reader? Phrases like "readers who loved X will find more to admire here" or "this will frustrate anyone hoping for Y" are signs the reviewer is thinking about you rather than themselves.
- Are they judging the book that exists, or the book they wanted? Complaint is legitimate when a novel breaks its own promises. It is less useful when it simply refuses to be a different genre.
Those signals help you use a review even when you suspect you would disagree with its verdict. A well-aimed three-star review often tells you more than a perfect five.
Disclosure, Free Copies and Affiliate Links
Reviews exist inside an economy, which is fine, as long as you can see it. In many countries, including the UK and the EU, reviewers are expected to disclose gifts, free review copies, sponsorship and affiliate links; rules vary by country, so treat visible transparency rather than legal detail as your practical guide. What you are looking for is simple: does the reviewer tell you how the book reached them?
A line like "I received a free copy from the publisher" is not a red flag. Hiding it is. The same goes for a page that earns a commission when you click through to a bookshop — perfectly normal, but it should say so, and the recommendation should read as if it would exist without the link. Watch for the tell-tale pattern: nothing but praise, no reservations at all, and a purchase button in every third paragraph.
Star ratings on retail sites deserve their own caution. A cluster of very short five-star reviews posted within weeks of release, sharing the same phrases, tells you about a campaign rather than a readership. Longer, mixed, slightly awkward reviews tend to be written by people who actually finished the book.
How Spoilers Are Handled
Spoiler handling is a small test of courtesy, and courtesy correlates with care. A reviewer who warns you before revealing a twist — "mild spoilers for chapter twelve below" — is thinking about your experience of the book. One who casually describes the ending in the second paragraph either has not considered it or does not care.
There is a middle ground worth recognising too. Some reviews reveal a plot point deliberately, because the reviewer believes you need it to judge whether the book is for you, and they flag it clearly so you can choose. That is not careless; it is a decision, announced. The distinction between "spoiled by accident" and "spoiled on purpose, with warning" is one of the fastest ways to judge whether a reviewer is reliable.
Signals of a Weak Review
Some warning signs are quick to spot once you know them:
- Plot summary with no argument. Two hundred words of what happens, one sentence of opinion.
- Attacks on readers rather than the book. "Anyone who enjoys this is kidding themselves" is not analysis.
- No sense of the book's ambitions. A review that judges a comic novel for not being solemn, or a quiet novel for not being thrilling, has missed the assignment.
- Uniform verdicts. A reviewer who adores everything with equal fervour offers no information; judgement requires discrimination.
- Copy-paste phrases. Identical wording across several reviews on the same site is worth noticing.
One of these alone proves nothing. Two or three together usually means you should read the next review instead.
A Quick Checklist Before You Trust a Review
Next time you are standing in a bookshop with your phone out, run through this in about thirty seconds. Does the review make at least one specific claim about craft, structure or style? Does it tell you who the book suits? Is the reviewer's own reading taste visible or inferable? Is any free copy, sponsorship or affiliate link disclosed? Are spoilers flagged? And does the verdict follow from the evidence given, or does it arrive from nowhere?
If most answers are yes, you have found something worth reading carefully — even if you end up disagreeing. Then read two more reviews of the same book from different sources. One opinion is a data point; three with visible reasoning form a pattern you can actually use. Over time you will find a handful of reviewers whose taste maps onto yours, and you will start to notice how they phrase a reservation, how they handle a book they admired but did not love, and how honestly they tell you when a novel was simply not for them. That is the real skill, and it accumulates. The stars are just the front door.
Photo: Towfiqu barbhuiya / Pexels
