Nobody sane builds a gas station without counting the cars that pass the corner. Authors skip that count all the time. They spend a year writing, upload the finished book, and only then go hunting for where it belongs on Amazon, where they discover the corner gets six cars a day or already has three stations on it. I hate that, because twenty or thirty minutes of research before the first chapter would have told them.
Count the cars first.
What does Amazon category research tell you before you write a book?
It tells you whether anybody is buying books like yours, how many copies the leaders move each month, how crowded the shelf is and how long the competing books run. All of that shapes the book before a word exists.
A category where the top title sells a few hundred copies a month is a different market from one where it sells thousands. You’ll write, price and position for each one differently, and you want to know which one you’re walking into while the outline can still change.
I told a client exactly that on a screen share once: “We’re not necessarily trying to nail down categories yet. We’re trying to look at the marketability and where it fits.” I told him we could spend hours and hours on it and we’d spend twenty or thirty minutes, max. Early research gives you the lay of the land, and the exact categories get locked in near publication, when the title, subtitle and description are final.
Page counts surprise people.
When you pull up a category, you see how thick the sellers are. If every book near the top runs 200 to 300 pages and your outline is heading for 600, you’ve learned something about your readers before you’ve written a page for them. Will your readers sit through twice the length they’re used to? Maybe the right answer is a shorter book. Maybe it’s a different category where the big books live. Either way, you’d rather find out now than after the editor’s invoice.
Reviews matter too. Read a few from each top book, good ones and bad ones, to see what readers praised and what they complained about. On that call I said I look at a couple from each. One-star reviews are mostly garbage, and some come from review bombing, where people pile onto a book because they don’t like the author. A pattern in the three-star reviews is gold, though. “States the obvious” or “badly organized” tells you exactly what your book has to do better than the books already on the shelf.
The two tools I use: KDSPY and Publisher Rocket
Amazon doesn’t publish sales numbers. It publishes a Best Sellers Rank for every book, and the rank moves with recent sales. Tools take those ranks and turn them into estimated sales.
As I told that client, “These are tools that let you look directly at Amazon. Amazon’s numbers from Amazon’s database.”
KDSPY is a browser extension. You open an Amazon search or a category page, click the extension, and it lists the top books with estimated monthly sales, estimated revenue, page count, price and review count. It also rates the category on popularity, potential and competition, with colors, so you can read the verdict at a glance. Green is friendly. Red is a warning, and you’d be a fool to ignore it.
Publisher Rocket is desktop software that used to be called KDP Rocket. Its keyword search shows estimated search volume. Its competition analyzer lists the books ranking for a keyword with their estimated sales, page counts and reviews, and its category search covers print editions as well as Kindle. I lean on it more as a book gets close to publication, when keywords and positioning matter most.
If you only buy one, my Publisher Rocket pick explains what you get for the money.
Neither tool is perfect. “It’s a nightmare, by the way, to figure out this stuff without this tool,” I said in that session, and a few minutes later the tool throttled me for running too many searches. Every once in a while Amazon changes something on its end and these products break until the developer catches up. I find that maddening. Still, I use them, because the alternative is guessing, and guessing with a year of your life is stupid.
Treat the sales figures as estimates. They’re calculated from rank, and a handful of sales can swing a rank and throw the monthly figure way off. The comparisons hold up better than the raw numbers.
If one category’s leader shows ten times the sales of another’s, the gap is real even when the exact figures are off.
How do you read monthly sales numbers for an Amazon category?
Look at three numbers: what the top book sells, what the book at the bottom of the first page sells, and how much money the whole page is making. The top gives you the ceiling. At the bottom, you learn what it takes to be seen at all, and the total tells you whether anybody is making a living there.
In one category I checked for a business book, the top title was selling about 239 copies a month. Further down, a book was selling 23. I called it “a low competition category” and said, “It’s not popular at all, because it only has 239 sales at the top.”
Then I told him why that was good news for him: “What it tells us is there’s a place where you can compete.”
A new book doesn’t need thousands of sales to show up on a page like that. It needs a few dozen.
Another category we checked had a top seller at 1,338 a month and a bottom seller at 35. Low competition again, with better potential. “You can get 35 sales pretty easily,” I said. “You’re not going to get a bestseller list from this particular one.” A category like that won’t make you famous. It will put your book in front of people who already shop for that subject, and a first-time author should want exactly that.
Watch the money column too, because you can get fooled there. A category can show decent sales and terrible revenue when everything near the top is priced at 99 cents. Well-priced hardcovers at the top tell you the readers there will pay for a serious book.
What does a red competition rating mean on KDSPY?
It means the leaders are selling thousands of copies a month, and a self-published book has almost no chance there.
Healthy competition is fine. I put it this way on that call: “Competition is healthy. That means you’re gonna have to do some work. If it’s red, and you’re self-published, don’t even bother. You’re not gonna make it.”
I still mean every word. One of the books I pulled up that day had more than 36,000 reviews, and I said, “That was selling pretty damn good.” Books at that level have traditional publishers behind them who send out review copies, buy placement and hire publicists. A lone author can’t do any of that without serious money for PR. Why would you pick a fight with a heavyweight when there’s an open ring next door?
A self-published book dropped into a red category sinks without a sound. It sits on page nine of a shelf nobody scrolls, it gets no reviews because nobody finds it, and the author decides the book was bad. Plenty of those books were fine. The author lost in the first round, and it cost them a year of work and most of their confidence. That makes me angry, because the warning was sitting right there in red.
If a category reads red and you’re self-published, don’t even bother. You’re not gonna make it. – Richard LoweShare on X
How many Amazon categories can a book have in 2026?
Three per format on KDP, as of 2026.
Amazon overhauled category selection in 2023. Before that, authors could message support and ask for extra categories, and plenty of them did. That door is closed. Good riddance. The trick mostly helped people cram books onto shelves where they didn’t belong.
When I walked a client through this in early 2024, I told him, “They only let you do three,” and said the old trick of asking support for more was getting rejected. Amazon can still list your book under other categories on its own, based on what customers do, and you don’t control those placements. Your three choices are the only ones you get to make, so make them count.
My own books reach Amazon a different way now. Amazon terminated my KDP account, and I publish through IngramSpark.
IngramSpark has you choose BISAC subject codes, the book trade’s standard subject list, and it recommends using three. Amazon places those books from that metadata. You don’t pick the Amazon browse category directly. So the research matters even more there, since the subject codes you choose are what point the book toward the shelves you’ve already scouted.
My self-publishing guide, Publish Your Book, covers the full setup for both routes. Do the research first either way.
One tough category and two you can win
My strategy for those three slots, in the words I used on that call: “Make one of them a really tough competing one and the other two to be easier. The other two pull up the other one.”
Those two easier categories are where your book can rank quickly. A few dozen sales in a low competition category can put you near the top of its bestseller list, and shoppers browsing that list will see you. Those sales count everywhere the book is listed, including the tough category, where they push your rank up a little at a time.
You won’t top the tough one with launch week sales from friends. Enough steady sales can lift you onto its first page. Real readers browse there.
Pick the tough category for the readers you want most, even if you’ll never lead it. Pick the easier two because they’re accurate and you can win them. Both still have to describe the book truthfully. A cookbook stuffed into a poetry category can rank for a day and then get returned by a reader who feels cheated, and Amazon has rules against miscategorized books anyway. Accurate and small beats inaccurate and empty every time.
When the search comes back empty
The research can come back with nothing at all. On that call we searched one narrow angle for the book in a keyword tool and got zero results.
“Didn’t find a damn thing,” I said. A second search came back empty too. “This is totally hit or miss, by the way.”
An empty result can mean two opposite things. Either you’ve found open ground nobody has written for, or you’ve found a subject nobody searches for. Your job is to figure out which. Check the keyword’s estimated search volume in Publisher Rocket, then search the broader subject one level up and see whether those books sell. Look at Amazon’s autocomplete suggestions too, since they come from real searches.
If people search for the broader subject and nobody has written your angle, you’ve got a gap worth filling. If nobody searches for any of it, you’ve learned that before spending a year on it. Isn’t that worth twenty minutes?
How do Amazon keywords affect a book title?
Keywords decide whether your title shows up when a shopper searches. In a later session I told a client that a book won’t sell without marketing, with a bad cover, or with the wrong title and description. A weak title “might not search right, might not have keywords in it.”
Paperback and hardcover titles are fixed once you publish. Changing one means a new edition and another ISBN, while a Kindle title can change whenever you like. Find your keywords before you name the book.
Keep them out of the description, though. Years ago I stuffed my descriptions with keywords on bad advice, and I explained what that cost me in why your book description isn’t converting. Amazon isn’t your only market either. Apple Books, Kobo and the stores reached through distributors like Draft2Digital all carry ebooks, and each of the self-publishing platforms has its own shelves. Amazon is where the data is easiest to get, so I start there.
Are Amazon bestseller guarantees worth paying for?
No. Some publishing packages promise a number one Amazon bestseller as part of the deal, and I went through one of those offers with a client. My verdict was two words: “Worthless guarantee.”
The trick takes almost no effort. “You just find a category that has the top bestseller in the category is like fifty books, and you just push it out through that,” I said. “You can get a number one within a few, easy.”
They run the same research I’ve described, aimed at the tiniest, most obscure category they can find. Then they push a burst of discounted sales through it, take a screenshot of the orange badge and sell you the title “bestselling author.” Nobody vets that badge. The whole racket insults readers and every author who earned the label, and I’ve written more about it in why bestseller lists mislead you.
I like to be real and know that it’s legit.
My book Focus on LinkedIn sold 15,000 copies in three days and reached the top of three categories. I priced it low and promoted the hell out of it, and as I told a client later, you can only keep that up for so long when you’re doing it all manually. A category list shows a book to shoppers only while something is pushing sales into it. Categories amplify demand you create, and they can’t create it for you. Do you want the badge, or do you want readers?
Twenty minutes now or a year of silence later
The whole process takes less time than people expect. Before you finalize the outline, open KDSPY on the three or four categories closest to your subject and write down the top sales figure, the bottom figure on the first page, the typical page count and the competition color for each.
Then run your main subject through Publisher Rocket’s keyword search, read a handful of reviews on the leading books and choose a working set of one tough category and two you can win. Revisit the set before publication, when the title is settled and the tools have fresher numbers.
You’ll know where the book can compete, how long it should run and what its title has to contain. You’ll also know when to walk away, because some ideas belong to traditional publishers with marketing budgets.
The publishing and marketing hub covers what happens after the research, and my self-published author service gets a finished book set up and positioned to sell with you. Writing a whole book and then discovering nobody buys anything like it is a damn waste of a year. Count the cars before you pour the concrete.
