How to Choose Comp Titles for Your Book

An author sent me her comp list last month and every title on it was a bestseller from 2012. She’d chosen them because she loved them, which is the most common way comps get picked and the least useful. A comp title has a job, and the job is to tell a specific person, in a specific place, what your book will feel like before they’ve read a page. Different places need different comps, and a book you adored fifteen years ago rarely does any of those jobs today.

So this is how I choose them for clients, where each kind gets used, and the one rule I hold to on every single one: the comp is a compliment, never a comparison you win.

What a comp is for

A reader who has just finished a book they loved is the easiest reader in the world to sell to, because they already know exactly what they want next: more of that. A comp borrows the feeling of the book they just closed and points it at yours. That’s the whole mechanism. It works on a stranger in seconds, it works on an agent scanning a pitch, and it works on Amazon’s ad system, which will happily show your book to people who bought the one you named.

It stops working the moment the promise breaks. If you comp a slow-burn small-town romance to a book that’s 90% banter and open door, the reader who arrives expecting that heat will tell everyone in the reviews that you misled them. Romance readers in particular hold you to the comp. Choose for accuracy first and popularity second, always.

The three tests a comp has to pass

Recent. Published in the last three to five years, so the reader you’re reaching still has it on their phone. A 2012 title tells an agent you haven’t read the market since 2012, and it tells Amazon nothing at all, because the people who bought it aren’t browsing anymore.

Reachable. Big enough that readers know it, small enough that the comparison is believable. Comping your debut to the biggest romantasy on the planet makes the reader laugh, and it makes the ad audience so wide that you’re paying to reach people who read one book a year. The sweet spot is an author with a devoted readership and several books, whose fans are hungry between releases. Those are the readers who take a chance on someone new.

Real. The comp shares the emotional shape of your book, and this is where most lists fail. Same subgenre is the floor. The comp needs the same heat, the same pace, the same relationship to darkness. A grumpy-sunshine sports romance with a happy ending and a grumpy-sunshine sports romance where the hero’s wife is alive in the epilogue are both sports romance, and they are not comps for each other.

When I’m doing this for a client I pull the also-boughts and the “customers also read” carousel on ten books in her subgenre, read the one-star and three-star reviews of each (they tell you what readers expected and didn’t get), and only then start a list. It takes an afternoon. The list usually ends up with eight names on it, sorted by which job each one does.

Comps for your metadata

This is where authors get themselves into trouble, so I’ll say the rule plainly: never put another author’s name or book title in your KDP keywords. Amazon’s own metadata guidelines prohibit it, and a listing that breaks that rule can be suppressed. The comp’s job in metadata is to give you language, never a name.

Go and read the comp’s listing. Look at which categories it sits in, what its subtitle promises, which trope words its blurb leans on and in what order. If your three closest comps all sit in Paranormal Romance and Vampires, and you’ve filed yourself under Gothic Fiction because you liked the sound of it, you have just found your first fix. Their categories tell you where your readers already browse. Their trope vocabulary tells you which words those readers type. Categories and keywords are their own subject, but the comp list is where the research for them starts.

Comps for pitches and blurbs

An agent or a bookseller wants the “X meets Y” line, and the version that works names two things the reader knows and produces something new in the space between them. “A Court of Thorns and Roses meets Ocean’s Eleven” tells a person exactly what to expect in seven words. “For fans of romantasy” tells them nothing they couldn’t get from the cover.

The pitch comp has room to play. One title can be a book and the other a film or show, and that combination often does the job better than two books, because a film comp carries tone the way a book title can’t. And you write the line so both halves are praised. “If you loved the found family in Fourth Wing, this is the book for you” sits well with the reader who loved Fourth Wing. “Like Fourth Wing but with better world-building” sits badly with everyone, including the reader you were trying to reach, who is a fan of that book and has just been told her taste is wrong.

Never knock the comp. The comp author is your peer, their readers are the audience you’re courting, and a put-down reads exactly as small as it is. Endorse them, then state your book as its own positive thing.

Comps for ads

Amazon Ads let you target a comp’s ASIN directly, which is the most literal use of a comp there is: your book appears on their product page and in the carousel under their listing. This is where the “reachable” test earns its keep. Targeting a title with 40,000 reviews puts you in an auction against everyone else who had the same idea, at a cost per click most indie budgets can’t sustain. Targeting the ten mid-list authors whose readers rate them consistently at 4.5 and who release twice a year gets you cheaper clicks from people more likely to buy an unfamiliar name.

On Meta, the comp turns into audience language rather than a target. You cannot name an author in the interest targeting the way you can on Amazon, and using another author’s name in the ad copy invites a report. The copy says “for readers who stayed up too late with their last enemies-to-lovers” and the image and headline do the rest. Facebook and Amazon ads work differently in enough ways that the same comp list produces two different campaigns.

Comps for A+ content

A+ content is the place you’d expect comps to shine and the place they’re least allowed. Amazon’s A+ guidelines reject modules that reference other books, authors or brands, so “for fans of” with a named title gets your submission sent back. What survives is the comp’s positioning without its name: the trope stack, the mood, the promise of the ending. If your comp research told you readers in your corner care most about the slow burn and the guaranteed happy ending, those are the two things the A+ modules say, in the reader’s own words, with the cover art carrying the rest. I’ve covered what goes in A+ modules and what gets rejected separately, and comps are the most common rejection I see.

Keep the list alive

A comp list built in 2024 is a 2024 document. Go back to it before every launch and every ad refresh, drop anything older than five years, and add the two or three books your readers keep mentioning in their reviews of you. Those unprompted mentions are the best comps you will ever find, because readers chose them, and readers are the only audience the comp was ever meant to convince.

If you’d rather hand the afternoon of also-bought scrolling to someone who does it weekly, the comp research is part of my book marketing strategy service: where your book sits, who is already reading beside it, and what to say to bring those readers across. Tell me what you’ve got and what feels messy.