How to Write a Romance Review People Will Use
A good romance review tells a stranger whether to spend her Saturday on this book, and the old site's rules were built for exactly that job.
Picture a review that reads, in full: “This book was AWESOME. I couldn’t put it down. Best one she has ever written. 10.”
The old site would not have posted it, and I would not post it now. The reviewer clearly loved the book, and there is nothing wrong with loving a book. But the post tells a reader nothing. No name for the heroine, no hint of the setting, no reason for the 10, and a capitalised word that breaks one of the plainest rules on the reviewer guidelines page. Now picture the same reviewer writing four paragraphs about a widowed schoolteacher in Montana and a railway surveyor who cannot stop measuring things. That is a review people will read.
The gap between those two posts is the whole craft.
The rules the site actually enforced
New reviewers sometimes assumed the guidelines were suggestions. They were not. Reviews without comments were not posted at all, and the site reserved the right to decline any review without notice. The core rules, as they appear in the restored archive:
- An overall rating from 1 to 10. The Young Adult site used 1 to 5 instead.
- A second rating specific to the subgenre: a History rating for historicals, a Paranormal rating for paranormals, and so on.
- A Sensuality rating, judged separately from how much sex the book contains.
- Comments required, written in 3 to 7 paragraphs, with a basic outline of the story and an introduction to the main characters.
- No writing in all capitals.
- As much detail as you like, without giving it all away.
- At least one review a month to stay on the active reviewer list; reviewers who missed a month were moved to inactive status until they posted again.
If you want to see how a submission travelled from your keyboard to the public page, how it works lays out the process.
Three numbers, three different questions
The site asked for three ratings because a single number blends too many things. A reader who wants a meticulous Regency and a reader who wants a fast sexy romp need different information from the same review.
The overall rating
The 1 to 10 overall score answers one question: how good is this book as a romance? Not as literature, not compared with the classics on your shelf from college. Within the genre and its promises, did it deliver?
A rough calibration I used with new volunteers, which you can take or leave:
| Score | What it means to a reader |
|---|---|
| 9–10 | You will buy her backlist tonight. Rare, and it should stay rare. |
| 7–8 | A good book. You will remember the hero next month. |
| 5–6 | Competent, forgettable, fine for a flight. |
| 3–4 | Real problems: a plot that falls apart, characters who change for no reason. |
| 1–2 | You finished it out of duty and want the reader to know why. |
The most common error in any review archive is score inflation. When half the reviews in a community sit at 9 or 10, the number stops carrying information. A 7 is a compliment. Write as if you believe that.
The subgenre rating
The second number asks how well the book does the specific job of its subgenre. For a historical, the History rating measures period texture and accuracy: does the heroine think like a woman of 1820 or like a 1999 office manager in a bonnet? For a paranormal, the Paranormal rating measures whether the world has rules and keeps them. A vampire who cannot cross running water in chapter two should not be paddling a canoe in chapter twenty.
A book can score 8 overall and 4 on History. That combination tells a reader something useful: fun book, do not pick it up for the period detail. If you are unsure what each subgenre promises, the field guide to romance subgenres covers how the communities drew their lines.
The sensuality rating
This rating caused more confusion than the other two combined, and the guidelines anticipated it with a line I have quoted to new reviewers more often than anything else in the archive: “often a book may have a great deal of sex, but may very well not actually be sensual.”
Sensuality is tension, attention to the body, the charge between two people in a room. Some books with six explicit scenes have none of it. The scenes read like an assembly manual. Other books keep the bedroom door shut and still make you aware of every time his hand brushes hers on a banister. The rating asks you to measure the charge, not the count.
How to describe heat level in words
The sensuality number is not enough on its own. Readers want to know what to expect, and the comments are where you tell them. I suggest covering four things, in plain language, in a sentence or two:
- Whether the love scenes happen on the page or behind a closed door.
- Roughly how many there are, and where they fall. Two scenes in the last third is a different book from five scenes starting in chapter three.
- How explicit the language is. Euphemism, frank description, or somewhere between.
- Anything a reader might want warning about, stated without judgement.
A sentence like “Two love scenes, both late in the book, both explicit, and the second one changes how he treats her the next morning” gives a reader everything she needs. It also tells her the scenes do work in the plot, which is the best compliment you can pay a love scene.
Avoid moralising. Some readers want closed-door books and some want the opposite, and both are right for themselves. Your job is accurate labelling.
Building the 3 to 7 paragraphs
The guidelines asked for a basic outline and character introductions. In practice, the reviews that readers bookmarked followed a shape something like this.
Paragraph one sets the scene. Where and when, who the heroine is, what she wants at the start. Keep it to the first few chapters.
Paragraph two introduces the hero and the obstacle. Romance runs on obstacles, and naming the central one (“she needs his land, he needs her silence about his brother”) tells a reader the shape of the book without spoiling it.
Paragraphs three and four cover what worked and what did not. This is where your opinion lives, and it needs evidence. A specific scene, a line of dialogue, a pattern you noticed.
The last paragraph gives a verdict and a recommendation. Who will enjoy this? Who should skip it?
That makes five paragraphs, which sits comfortably inside the 3 to 7 range. A slim category novel can carry three. A 500-page historical saga with two timelines might need seven. Past seven, you are writing an essay, and readers browsing the reviews on a lunch break will scroll past it.
Spoiler discipline
Romance readers know how the book ends. The couple gets together. That is the contract, and you can mention it without guilt. What you protect is everything else.
A few working rules:
Stay inside the first third for plot summary. Anything that happens after the midpoint should appear in your review only as a judgement (“the last act drags”), never as an event.
Describe the obstacle, not the resolution. “He has a secret about the night his father died” is fine. What the secret is, is not.
Treat suspense subplots as mysteries. In romantic suspense the villain’s identity is the ending, as much as the wedding is. Do not hint, even through your choice of praise. “I suspected the brother from his first scene” spoils the book as surely as naming him.
Watch your character introductions. If the heroine’s long-lost sister shows up in chapter twenty, she does not belong in your list of characters, however much you liked her.
Be careful with series epilogues. If book four ends with a pregnancy announcement for the couple from book one, that is a spoiler for anyone still reading book one.
When you cannot explain your criticism without giving something away, say so and keep it vague: “The resolution of the inheritance plot relies on a coincidence I could not accept.” Readers trust you more for the restraint.
Weak sentences and stronger ones
Most weak reviews fail at the sentence level. They state a feeling without the evidence that would let a reader decide whether she shares it. A few swaps of the kind I would suggest, with invented examples:
| Weak | Stronger |
|---|---|
| I loved the heroine. | She corrects the vicar’s Latin in her second scene, and I trusted her from then on. |
| The middle was slow. | Chapters nine to fourteen replay the same quarrel about the estate three times. |
| The history was bad. | A heroine in 1816 rides alone from Bath to London overnight and nobody remarks on it. |
| The sex was hot. | Three scenes, open door, frank language, and each one moves them closer. |
| Don’t waste your money. | I could not believe the hero’s reversal in the final chapters, and the book asks you to. |
The stronger versions do two things the weak ones skip. They show the reader a piece of the book, and they let her judge for herself. A reader who does not care about Regency travel customs can read that third example and decide the problem will not bother her. That is a successful review, even though she disagrees with you.
Reviewing a book you disliked
The hardest reviews to write are the bad ones. You worry about the author reading it. You worry about being mean. Some reviewers dodge the problem by never posting anything below a 6, which gives a site an inflated look and makes their good reviews worth less.
I don’t buy the argument that a negative review harms anyone except a careless author. A reader who spends money and a weekend on a book deserves to hear from someone who did the same and regretted it. The trick is to keep the review about the book.
Start with what the book was trying to do. A light contemporary romp should be judged as a romp, not faulted for lacking the depth of a family saga. Then explain where it fell short, with the same specificity you would use for praise. Name the scenes. Quote a line if it helps. Avoid any sentence about the author as a person, her motives, or her career.
Close with who might still enjoy it. Almost every book has a reader. “If you love marriage-of-convenience plots and do not mind a passive heroine, you may like this more than I did” is honest and generous, and it helps the right person find the book.
One more thing. If a book made you angry, wait a day before writing. Angry reviews read as angry, and readers discount them.
Series books need their own approach
Series romance boomed during the site’s life. Nora Roberts finished trilogy after trilogy, J.D. Robb’s In Death books arrived twice a year, and Julia Quinn’s Bridgerton family ran to eight novels between 2000 and 2006. Reviewing book five of anything raises questions that a standalone does not.
Say whether the book stands alone. This matters more to readers than almost anything else in a series review. Can someone start here, or will she be lost in a crowd of siblings and their spouses?
Judge the couple, not the franchise. Each romance novel in a series has its own hero and heroine. A beloved family cannot rescue a weak central pair, and your score should reflect the pair.
Note returning characters lightly. Readers of the series want to know their favourites appear. New readers do not need a census.
Check the author’s page. The archive’s series lists and author pages such as the one for Nora Roberts show reading order, and readers will thank you for pointing out that book three is the one to start with.
Before you post
Read the review out loud, or at least in your head at the speed you would speak. You will catch the problems a fast skim misses.
The most common ones in the archive are mechanical. Sentences that run on for fifty words because the reviewer kept adding clauses about the hero’s backstory. Pronouns that lose their owners, so a reader cannot tell whether “she” in the third sentence is the heroine, her sister or the housekeeper. Paragraphs that start with “This book” four times in a row. If you have trouble spotting these in your own draft, running it through a free clarity checker will flag the overlong sentences and the vague pronouns so you can fix them before a volunteer has to send the review back.
Then check the practical details. Title and author spelled correctly, because a misspelled name breaks the links to the author’s booklist. Publisher and year right. Your three ratings filled in, and consistent with your text: a review that spends four paragraphs on the book’s flaws and then awards an 8 confuses everyone.
Finally, check for capitals. The no-all-caps rule exists because capitalised words read as shouting on a screen, and a review that shouts loses the reader’s trust in the first line.
Why the one-a-month rule existed
Some reviewers grumbled about the requirement to post at least one review a month to stay active. It seemed bureaucratic for a volunteer site.
The rule had a reason. The site promised readers fresh reviews across every community, and the volunteers who coordinated each community needed to know who they could count on when the new releases landed. A list of a hundred reviewers with sixty of them silent was worse than a list of forty who posted. The monthly rule kept the list honest.
It also kept reviewers sharp. Writing one review a month is enough practice that you get faster and better at it. Writing one a year means starting from scratch each time, staring at a blank comment box and wondering how to describe a sensuality level without blushing.
If you want to pick it up again now, on the restored archive, start with a book you finished in the last week, while you still remember the heroine’s name.
Sources and further reading
- The 1 to 10 scale, the subgenre rating, the sensuality line quoted above, the 3 to 7 paragraph rule, the ban on all capitals and the one-review-a-month requirement: the site’s reviewer guidelines, archived June 2006.
- The eight Bridgerton novels, from The Duke and I (2000) to On the Way to the Wedding (2006): Wikipedia, Julia Quinn.
- The genre’s promise of an “emotionally satisfying and optimistic ending”, which is why the happy ending is no spoiler: Wikipedia, Romance Writers of America.