Romance readers buy by trope, subgenre and author, and many read through Kindle Unlimited. So romance ads work best when they target specific tropes and comparable authors, and when you judge them by total earnings from the whole series, including page reads, not by the ACOS on book one.
Last reviewed by the InteliAds team.
How romance readers search
A romance reader rarely types "romance novel". They type what they want to feel: enemies to lovers, grumpy sunshine, second chance, fake dating, small town, billionaire, sports, why choose, dark romance. Heat level and subgenre matter as much as trope. A reader of closed-door sweet romance will not buy a spicy dark romance, and the reverse.
That gives you a clear keyword strategy: build lists by trope and subgenre, and only include tropes your book actually delivers. Matching the reader's promise is what turns a click into a sale or borrow; a mismatch costs the click and the review.
Trope keywords: "enemies to lovers romance", "fake dating romance", "single dad romance".
Subgenre and setting: "small town romance", "hockey romance", "regency romance".
Comparable authors in the same heat level and subgenre.
Product targets: books whose covers, tropes and heat level look like yours.
Kindle Unlimited changes the maths
Romance is one of the genres where KU borrowing is common. Amazon Ads reports purchases, not page reads, so a romance campaign can show a high ACOS while the book is actually making money from KENP reads in your KDP report.
The fix is to measure differently. Each month, take total KDP earnings for the series (sales royalties plus page-read earnings) and compare it with total ad spend. If you want a per-click view, divide page-read earnings in the period by ad clicks to get a rough page-read value per click, and add it to the sales side when you judge a keyword. It is an estimate, but it is much closer to the truth than ACOS alone.
In a series, the ad usually belongs on book one. A reader who finishes book one and likes it moves on to book two without another ad click. That later income is why romance authors can run ads above book one's break-even ACOS and still profit.
Measure your read-through rather than guessing it: compare units (or KENP) for book two with book one over the same months. If 40 of every 100 book-one readers go on to book two in your data, you can value each book-one sale by its own royalty plus 0.4 of book two's, and so on through the series. Use your own numbers; read-through varies a lot by book.
Amazon's book ad moderation guide says covers cannot contain sexually suggestive content and gives examples such as undressing, lingerie and provocative embraces. Romance covers are where this most often bites. If an ad is rejected, the reason is shown in the console, and the guide says you may need to change the cover image used.
Before investing in campaigns for a spicy title, check that at least one ad is approved. Some authors use a discreet or object cover edition for this reason. Also make sure the book sits in the subgenre categories its readers browse, since category pages are where product targeting and organic visibility meet.
Where InteliAds helps romance authors
InteliAds brings in KDP royalties, including page-read earnings, through the KDP Sync Chrome extension, and lines them up with Amazon Ads spend per book. That is the view romance authors need: profit after ads per title and across a series, not ACOS alone.
BidBot is the automated bidding in InteliAds. You pick a preset (Suggest only, Safe, Balanced, Aggressive or Aggressive+) and a target ACOS between 10% and 80%. Safe waits for 20 clicks before it raises a bid and leaves 48 hours between changes to the same target; Balanced waits for 10 clicks and 24 hours. You can also set it to lower bids only, and mark promo windows when it should not raise bids.
For a KU-heavy romance, set the target ACOS above book one's sales-only break-even if your page-read and read-through data support it, and start BidBot in Safe or Suggest only so you can see its choices before they go live.
Check that your cover passes ad moderation by launching one small campaign first.
List the tropes, heat level and subgenre your book truly delivers.
Build one keyword campaign per main trope group (for example sports romance, small town).
Build one product targeting campaign on 20 to 40 comparable books with matching covers and heat level.
Run a small auto campaign to find trope phrases readers use that you missed.
Each month, compare total series KDP earnings (sales plus page reads) with total ad spend.
Set your target ACOS from that combined view, then let Rules or BidBot hold bids to it.
Worked example (illustrative numbers)
All figures are invented to show the method. Book one of a small-town romance series sells for $3.99 with a royalty of about $2.70, so its sales-only break-even ACOS is about 68%. It is in KU.
Last month, ads on book one spent $300 and Amazon Ads reported $330 in sales: 91% ACOS, which looks like a loss. The KDP report for the series shows $420 of page-read earnings and $180 of sales royalties on books two and three. Total series KDP earnings were about $825 (including about $225 royalty on the ad-attributed book-one sales) against $300 of ad spend. The series is profitable, so the author keeps the campaign and instead cuts the "billionaire romance" keyword group, which spent $70 with one sale because the book has no billionaire.
Frequently asked questions
Why is my romance ACOS so high?
Often because KU borrows and series read-through are not in ACOS. Compare spend with total series earnings, including page reads, before cutting.
Why was my romance ad rejected?
Amazon's moderation guide does not allow sexually suggestive covers. Check the rejection reason in the console and consider a different cover image.
Should I advertise every book in my series?
Start with book one. Ads on later books mainly reach readers who have not read book one, who are less likely to buy.
Which keywords work for romance ads?
Specific tropes, subgenres and comparable authors that match your book's heat level. Generic terms like "romance novel" tend to spend without converting.
Amazon and KDP rules change; check the linked help pages before making decisions. Worked examples use illustrative numbers, not results from real accounts.