The KDP listing: keywords, categories and a description that sells

Most short books that don't sell aren't bad books. They're invisible ones. The manuscript gets months of attention and the listing gets twenty minutes the night before upload, and then nobody searching Amazon ever sees it.

The listing is three things: the words Amazon matches searches against, the shelves it puts the book on, and the paragraph that decides whether someone who lands on the page buys. Each one is worth an hour. Here's how I'd spend it.

Keywords: seven boxes, used as phrases

KDP gives you seven keyword boxes. The most common mistake is treating them like hashtags: one word per box, often words already in the title. That wastes most of the space.

Think of each box as a phrase a buyer would actually type. Not "pricing" but "how to raise freelance rates". Not "design" but "pricing guide for graphic designers". Amazon already indexes your title and subtitle, so repeating those words in the boxes adds nothing. Spend the boxes on what the title doesn't say.

A quick way to find real phrases: start typing your topic into Amazon's search bar in the Kindle Store and write down what it suggests. Those suggestions come from what people search for. Then look at the top few books for each phrase. If the results are all big-name titles with thousands of reviews, pick a narrower phrase. A small book wins small searches first.

Things to leave out: other authors' names, other books' titles, words like "bestseller" or "free", and anything that isn't true of your book. Amazon's guidelines prohibit these, and they're the fastest way to get a listing suppressed.

Categories: choose the shelf you can stand out on

When you publish, you choose up to three categories. The instinct is to pick the biggest, most obvious one, like Business & Money. That's the shelf where your book is one of tens of thousands.

A short, specific book does better on a specific shelf. A guide for freelance designers belongs closer to something like freelancing or small business than to general business. Browse the category on Amazon before you choose it and look at the books that rank there. If they're the same kind of book as yours, and their sales rank isn't out of reach, that's your shelf.

Use all three slots, and make them different. Three nearly identical categories put you in front of the same buyers three times. A niche category, a related one, and one broader shelf covers more ground.

The description: written for the person who's almost convinced

Whoever reads your description has already clicked on the cover and title. They're interested. The description's only job is to answer "is this for me, and what will I get?" quickly enough that they don't go back to the search results.

What works, in order:

1. One line naming the problem, in the reader's words. "You know you're undercharging. You're afraid that if you raise your rates, your best clients will leave."
2. What the book does about it, concretely. Not "a comprehensive guide to pricing" but what someone can do after reading it: write the rate-increase email, handle the pushback, decide which clients to let go.
3. A short list of what's inside. Three to five specific points. Chapter titles often work if they're good.
4. Who it's for, and who it isn't for. This sounds like it would cost sales. It builds trust, and the people it turns away would have left a bad review.

Keep paragraphs short. Most people read Amazon on a phone, where a long paragraph becomes a wall. And don't open with praise for yourself or the book. Nobody believes a description that calls its own book "essential".

The part to be honest about

KDP asks whether your book contains AI-generated content. If you wrote it with a tool like this one, the answer is yes. That answer goes to Amazon, not onto your store page, so it doesn't change how buyers see the book. Answer it truthfully. A useful book doesn't need to hide how it was made.

Where GetPressReady fits

The publish kit drafts all of this from your finished book: a store description, seven keyword phrases, category suggestions, an author bio and a few alternative subtitles. I wrote about where it sits in the whole process in a one-prompt book, start to finish.

Treat those drafts the way you'd treat a capable assistant's work. They give you a strong first version in seconds instead of a blank box at midnight. Then you check the keywords against real Amazon searches and the categories against real shelves, because that's the part that depends on what's selling this week.

If you want to judge the writing first, the sample books are complete and unedited. Or start your own: the plan and the first chapter are free, and pricing for the rest is one page.