Your Keyword Strategy Is Probably Built on Rules That No Longer Exist
Here is what Amazon changed, why most erotica metadata advice is outdated, and the complete system for fixing it.
You did everything you were told to do.
You researched keywords. You looked at what was ranking. You filled your keyword slots with phrases that seemed right. You picked categories. You wrote a description. You published.
And then not much happened.
If you went looking for answers, you probably found the same advice recycled everywhere: check what the bestsellers in your niche are using, match your keywords to reader search behavior, use Publisher Rocket, fill all your slots. The advice was presented like a checklist. When it did not work, the implication was that you had not followed it carefully enough.
But here is the thing nobody in those guides mentioned.
Most of that advice was written to describe how Amazon’s algorithm worked in 2019 and 2020. The algorithm has had two major updates since then. The one that matters most is called A10, and it changed the rules in ways that make a lot of the old keyword advice not just ineffective but actively counterproductive.
What Changed With A10
The old system, A9, was a keyword matching machine. If a reader searched for a phrase that appeared in your metadata, you got a relevance point. Enough relevance points plus enough recent sales meant you ranked. The whole game was packing your metadata with the right terms and generating early sales velocity.
A10 still cares about keyword relevance. But it added three things that most authors do not know about.
The first is behavioral signals. A10 tracks what readers do after they find your book. A reader who lands on your page and immediately leaves is a negative signal. A reader who reads your description, scrolls through your content, and then buys is a strong positive signal. This means your blurb is now a ranking factor, not just a sales tool. If readers are bouncing off your page, it feeds back into your discovery performance.
The second is external traffic weighting. When someone clicks a link in your newsletter and lands on your Amazon product page, Amazon registers that as quality off-platform traffic. Every newsletter launch you run is contributing to your algorithmic ranking beyond the direct sales it generates. This is a genuine advantage for authors who have built a reader list, and it is largely invisible to authors who have not.
The third change is the one that does the most hidden damage, and it is the one almost nobody has written about for erotica authors specifically.
The Metadata Coherence Problem
A10 cross-references your metadata fields against each other. It is not evaluating your title, keywords, categories, and description individually anymore. It is checking whether they are all telling the same story about the same book.
Amazon calls this semantic coherence. When your fields are aligned, your relevance scores go up across all of them. When they are contradicting each other, even subtly, your rankings take a hit you cannot diagnose because Amazon never tells you what caused it.
This happens to erotica authors constantly. A book gets keyworded for one subgenre but categorized in another because the author is trying to avoid the adult section. The title implies one heat level but the description describes another. The keywords say dark romance but the categories say contemporary fiction. All of those choices were individually reasonable. As a system, they are incoherent, and the algorithm treats them that way.
You lose ranking in both places. Not because the book is bad. Not because the keywords are wrong. Because the metadata fields are pointing in different directions.
The Erotica-Specific Problem on Top of All That
Mainstream romance authors dealing with A10 have a fallback that you do not. They can run Amazon Ads to compensate for weak organic performance. You cannot. Amazon bans erotica and adult romance from its advertising platform. Your organic keyword relevance has to do the work that paid traffic does for everyone else.
That means metadata mistakes cost you more than they cost other authors. A confusing blurb, a weak keyword slot, a category mismatch, a coherence problem you did not know existed. For a romance author running ads, those are inefficiencies. For you, they are the difference between visibility and silence.
And then there is the adult dungeon.
If you have published erotica for any length of time, you know the term. What most authors do not know is that dungeon suppression and standard adult content flagging are two completely different things. Standard adult flagging is basically unavoidable for explicit content. Amazon scans your manuscript and tags it. That is normal. What it does is limit your book to readers who have enabled adult content in their account settings.
Dungeon suppression is different. It removes your book from search results, category charts, and recommendation widgets entirely. Readers who have adult content enabled cannot find it. Your own author page does not surface it properly. And Amazon sends you no notification that it happened. You just watch your sales collapse and spend weeks trying to figure out why.
The dungeon is almost always triggered by a metadata problem, not a content problem. There are specific keyword and category combinations that flag the suppression system. There are cover elements that trigger it. There are description patterns that are perfectly legal to write but that tell Amazon’s automated system something it does not like. Most of these triggers are not documented anywhere officially. They have been mapped by the erotica author community through years of painful trial and error.
That map is in this guide.
Erotica Keyword Domination
The Amazon Metadata and A10 Algorithm Playbook
A complete system for erotica and adult romance authors: keyword research, 10-slot strategy, adult dungeon avoidance, blurb formulas, category hacking, and metadata coherence.
What Is Inside
The guide is divided into eight parts, each one building on the last.
Part One covers the algorithm. You get a plain-language explanation of what A10 actually changed, why those changes matter more for erotica authors than for any other fiction category, and how the adult flagging system works at a mechanical level. If you have ever had a book disappear from search without explanation, this section explains why.
Part Two is keyword research built specifically for adult fiction. Most keyword guides assume you can run ads to test performance. You cannot. This section covers how to use the KDP search bar and Publisher Rocket to find what erotica readers are actually searching for, how to read competition density in niches where Amazon suppresses the bestseller data, and how to build a pool of 40 to 60 keyword candidates before you touch your dashboard. The research process for erotica is different from the research process for clean fiction, and this section treats it that way.
Part Three is the 10-slot strategy. Amazon quietly expanded from 7 keyword slots to 10 and most authors are still leaving three slots empty or filling them with redundant terms. This part covers a slot-by-slot allocation framework with worked examples, a phrase construction method that gets you maximum coverage across subgenre, trope, character dynamic, and heat level, and the one rule about keyword slots that most authors break without knowing it.
Part Four is adult dungeon avoidance. This is the section authors come back to most. It covers the specific keyword and category combinations that trigger suppression, the cover and description patterns to avoid, and a pre-upload checklist you run through before every new title goes live. It also covers recovery: what to do if a book is already in the dungeon, in what order to change things, and how long each fix typically takes to show results.
Part Five is category strategy. The difference between BISAC codes and Amazon browse nodes confuses most authors and costs them real discoverability. This part explains how the category system actually works, gives you a method for finding low-competition categories where erotica can rank without competing against titles from major publishers, and covers the decision of whether to go into adult-designated categories or attempt to rank in mainstream romance.
Part Six is the blurb. The standard blurb formula that works for literary fiction and clean romance does not work for erotica. Erotica readers are making a different kind of buying decision and they need different information presented in a different order. This part presents the four-part erotica blurb formula with five worked examples across different subgenres, including monster romance, dark romance, contemporary, BDSM, and reverse harem.
Part Seven is metadata coherence. This is where everything comes together. Chapter Seventeen explains what semantic coherence actually means in concrete terms and how to tell whether your metadata is coherent or not. Chapter Eighteen walks through a field-by-field metadata audit you can run on your existing catalog to find and fix the incoherence problems that may have been quietly suppressing your rankings for months or years.
Part Eight covers A+ content. Amazon limits what adult-flagged books can show on their product pages, but there is still meaningful space to work with. This part explains the limits, what you can actually do within them, and presents three A+ content templates you can adapt for your own books.
The Workbook
At the back of the guide is a set of working documents you use alongside the reading. These are not summaries of what you just read. They are the actual tools.
The keyword research workbook walks you through building your candidate pool step by step and gives you a place to record and score each candidate before you make final decisions. The 10-slot planning grid lets you map your final keyword selections against your title, subtitle, and categories so you can see coherence problems before they go live. The metadata coherence audit worksheet is a field-by-field checklist for evaluating any book in your catalog. The adult dungeon avoidance checklist is the pre-upload review that takes five minutes and has saved authors from suppression problems they would not have caught otherwise.
Who This Is For
This guide is for erotica and adult romance authors who are publishing on Amazon KDP, whether exclusive through Kindle Unlimited or wide.
It is for authors who have published at least one or two books and are not seeing the discovery results they expected. Not because the books are bad. Because the metadata is doing less work than it could.
It is for authors who have used keyword advice from general indie publishing guides and found that it does not quite translate to adult fiction. The erotica context changes too many variables for generic metadata advice to be fully applicable.
It is for authors who are afraid of the adult dungeon and have been making metadata decisions based on fear rather than information. Fear-based metadata decisions are usually wrong. Informed ones are much better.
It is for authors with a backlist who want to audit their existing titles and recover rankings they may have been quietly losing without knowing why.
It is probably not the right starting point if you have not published anything yet. The concepts will make more sense with at least one book in the market. If you are still writing your first book, save this for when you are ready to publish.
What You Will Be Able to Do After Reading It
You will be able to do keyword research for adult fiction using methods that account for the specific ways Amazon handles erotica search data.
You will be able to build a 10-slot keyword strategy that covers subgenre, trope, dynamic, and heat level without creating the metadata contradictions that suppress rankings.
You will be able to run the pre-upload checklist on every new title and know, before you publish, whether your metadata has any of the patterns that trigger dungeon suppression.
You will be able to audit your existing catalog and identify the coherence problems that may be costing you discoverability right now.
You will be able to write a blurb that works for erotica readers specifically, using a formula built around how adult fiction readers actually make buying decisions.
You will understand why your metadata functions as a system, not a collection of individual fields, and you will know how to make sure all the parts of that system are pointing in the same direction.
Erotica Keyword Domination: The Amazon Metadata and A10 Algorithm Playbook
Includes the full guide plus keyword research workbook, 10-slot planning grid, metadata coherence audit worksheet, and adult dungeon avoidance checklist.
Amazon is not going to tell you that your metadata is incoherent. It is not going to explain what triggered your suppression or why your book stopped ranking after a strong first week. It is not going to send you a note saying that three of your keyword slots are actively working against the other seven.
It is going to keep running its algorithm and your book is going to keep performing at whatever level your current metadata allows.
That level can be higher. If you have published a well-written book with a good cover into a category where readers are actively searching, the gap between where it is ranking and where it could rank is almost always a metadata problem. Fixable. Documented. This guide shows you how.