Last updated: December 2024

KDP Keyword Research: Why I Stopped Guessing and Finally Started Making Sales

I published my first 47 books on Amazon KDP without doing proper keyword research. The result? €412 total. Then I learned how KDP keyword research actually works—and my next 50 books made €3,200. Here's everything I discovered.

The Real Problem: You're Writing Books Nobody's Searching For
What KDP Keywords Actually Do (And Don't Do)
My Step-by-Step Keyword Research Process
The Tools I Actually Use (And One I Stopped Using)
The Mistakes That Cost Me Money
Putting It All Together: A Real Example
The Honest Truth About Keyword Research
What To Do Next

The Real Problem: You're Writing Books Nobody's Searching For

Most beginner KDP publishers pick keywords the same way I did—gut feeling.

"Dark romance billionaire sounds good."

"People love dragons, right?"

"This title sounds catchy."

That's not keyword research. That's hoping.

The problem is that Amazon KDP gives you exactly seven keyword slots. Seven chances to tell Amazon's algorithm who should see your book. If you fill those slots with random phrases, you're essentially hiding your book in a warehouse with 15 million other titles.

I made this mistake for months. I'd spend 60 minutes generating a novel with WriteAIBook.com, another 30 minutes editing, maybe an hour on the cover—and then 3 minutes throwing random keywords into the KDP dashboard.

That 3 minutes of laziness was costing me hundreds of euros per book.

What KDP Keywords Actually Do (And Don't Do)

Before diving into tactics, let's clear up a misconception that cost me money.

KDP keywords don't directly boost your rankings. They tell Amazon which searches your book should appear in.

Think of it this way: if someone searches "enemies to lovers dark romance," Amazon scans millions of books to find relevant matches. Your seven backend keywords, plus your title, subtitle, and description, determine whether your book shows up in that search.

But here's what most guides don't tell you: showing up isn't enough. You need to show up for searches where buyers actually click and purchase.

I had one book ranking on page one for "paranormal romance werewolf alpha." Sounds great, right? Except that search gets maybe 200 searches per month, and the people searching it have very specific expectations my book didn't meet. Lots of impressions, almost zero sales.

Meanwhile, another book ranked on page three for "dark romance kidnapping" and outsold it 8 to 1. Fewer impressions, but the buyers were actually looking for what I'd written.

This is why kdp keyword research matters more than raw search volume.

My Step-by-Step Keyword Research Process

After testing this across 350 books over six months, here's the exact process I use now. It takes about 20 minutes per book—and it's made a measurable difference in my results.

Step 1: Start With Amazon's Search Bar (Free)

Open Amazon, go to the Kindle Store, and start typing. If I'm publishing a dark romance, I'll type "dark romance" and watch what Amazon suggests. These autocomplete suggestions are gold—they're based on what real people actually search for.

I screenshot everything. "Dark romance mafia." "Dark romance enemies to lovers." "Dark romance bully." "Dark romance kidnapping."

Then I go deeper. I type "dark romance m" and see what comes up. "Dark romance morally grey." "Dark romance motorcycle club." "Dark romance mafia arranged marriage."

I spend 10 minutes just collecting these suggestions. Usually I end up with 30-40 phrases.

Step 2: Validate With Competitor Analysis

Next, I find the top 10 books in my target category that are actually selling.

Here's how to spot them: look at the Best Sellers Rank (BSR). Anything under 50,000 in the Kindle Store is selling at least a few copies daily. Under 10,000 means it's doing really well.

I open each book's page and look at:

The title (what keywords did they include?)
The subtitle (more keywords)
The "Customers also bought" section (related keywords)
The categories they're listed in

I'm not copying their keywords. I'm understanding what's working in this market.

Step 3: Use Publisher Rocket (Paid, But Worth It)

I resisted paying for tools for months. That was dumb.

Publisher Rocket costs around $100 one-time and has paid for itself many times over. Here's what it does that free methods can't:

Shows estimated monthly searches for specific keywords
Shows competition scores (how hard it is to rank)
Reveals what categories competitors are in
Suggests related keywords you'd never think of

I plug in my top 10 keywords from steps 1 and 2, and Publisher Rocket shows me which ones have actual search volume versus which ones just sound good.

The difference was stark:

Books with researched keywords averaged €67 in their first 90 days.

Books with gut-feeling keywords averaged €18.

Step 4: Check the Competition Reality

Search volume means nothing if you can't compete.

If I find a keyword with 5,000 monthly searches but the top 10 results are all books with 500+ reviews and BSRs under 1,000, I'm not going to rank there anytime soon.

I look for what I call "opportunity keywords"—decent search volume (500+ monthly) with beatable competition (top results have under 100 reviews, or BSRs over 20,000). These are the keywords where a new book can actually get visibility.

Step 5: Fill Your Seven Slots Strategically

Here's my framework for the seven keyword slots:

Primary keyword phrase (highest volume, most relevant)
Secondary keyword phrase (related but different angle)
Reader trope/expectation (enemies to lovers, forced proximity, etc.)
Setting or subgenre modifier (mafia, small town, paranormal)
Emotional hook (angst, slow burn, dark)
Comparable author or series style (without using actual names)
Long-tail specific phrase (3-4 words, very targeted)

Don't waste slots on single words. "Romance" is useless—it's in your category already. "Enemies to lovers dark romance mafia" is much more useful as a single keyword phrase.

The Tools I Actually Use (And One I Stopped Using)

Publisher Rocket: My Primary Tool

I mentioned this above, but it's worth emphasizing. For amazon keywords kdp research, this is the most practical tool I've found.

It's not perfect. The search volume estimates are approximations, not exact numbers. But the directional data is solid enough to make better decisions.

KDSPY: Useful for Category Research

KDSPY is a Chrome extension that shows BSR history, estimated earnings, and category data right on Amazon search results. I use it mainly to validate whether a niche is actually profitable before I commit to writing a series in it.

Amazon Ads Keyword Tool: Free and Underrated

If you run Amazon ads (and you should, eventually), the advertising console has a keyword suggestion tool that's surprisingly useful for organic research too. It shows you related keywords, suggested bids (which indicate competition), and search term reports from your actual ads.

What I Stopped Using: Generic SEO Tools

I wasted time with Ahrefs and SEMrush trying to research Amazon keywords. These tools are built for Google, not Amazon.

Amazon's search algorithm is completely different. People searching on Amazon have purchase intent. People searching on Google might just want information. Use Amazon-specific tools for Amazon research.

The Mistakes That Cost Me Money

Mistake 1: Stuffing Keywords Into Titles

I tried putting "Dark Romance Mafia Enemies to Lovers Kidnapping" as a subtitle once.

It looked spammy. It read terribly. And Amazon's algorithm is smart enough to penalize keyword stuffing.

Your title should be readable and intriguing. Save the keyword optimization for your backend slots.

Mistake 2: Ignoring Category Selection

Keywords get you found in search. Categories get you found in browse.

I had books with perfect keywords that still flopped because I put them in the wrong categories. A dark romance in "Contemporary Fiction > Women" is competing with literary fiction readers who don't want what I'm selling.

Spend time researching categories as carefully as keywords. Publisher Rocket helps here too.

Mistake 3: Never Updating Keywords

Amazon lets you change your keywords anytime. Most publishers set them once and forget.

I go back every 90 days and update keywords on underperforming books. Sometimes a small tweak—changing "billionaire romance" to "billionaire boss romance"—makes a noticeable difference.

Mistake 4: Copying Competitor Keywords Exactly

If the top book in your niche is using "alpha werewolf fated mates," that keyword is already dominated. You're not going to outrank a book with 2,000 reviews.

Find related keywords they're not targeting. Look for angles they missed.

Mistake 5: Prioritizing Volume Over Intent

A keyword with 10,000 monthly searches sounds better than one with 500. But if those 10,000 searchers are looking for free books or something different than what you wrote, you won't convert.

I'd rather rank #3 for a 500-search keyword with high purchase intent than rank #15 for a 10,000-search keyword where nobody buys.

Putting It All Together: A Real Example

Let me walk you through a recent book.

I was publishing a dark romance with a mafia hero, arranged marriage trope, and enemies-to-lovers arc. Using WriteAIBook.com, I generated the novel in about an hour, then spent another 30 minutes editing.

For keyword research, I:

Typed variations into Amazon's search bar and collected 35 suggestions
Analyzed the top 8 selling books in dark romance mafia
Ran my top keywords through Publisher Rocket
Found that "arranged marriage mafia romance" had 2,400 monthly searches with moderate competition
Discovered "dark romance Italian mafia" was less competitive than "dark romance mafia"

My final seven keywords:

arranged marriage mafia romance
dark romance italian mafia
enemies to lovers dark romance
possessive hero romance
forced marriage romance dark
mafia romance series kindle unlimited
morally grey hero romance

The book hit #45,000 BSR within two weeks and has made €89 in its first 60 days. Not a home run, but solid for a book I spent less than two hours creating.

The Honest Truth About Keyword Research

Here's what I wish someone had told me at the start:

Keyword research improves your odds. It doesn't guarantee success.

I've had perfectly researched books flop and poorly researched books take off. The algorithm has randomness baked in. Reader tastes shift. Competition changes daily.

But across 350 books, the pattern is clear: books with researched keywords consistently outperform books without.

It's like poker. You can play perfect strategy and still lose individual hands. But over hundreds of hands, good strategy wins.

What To Do Next

If you're just starting with KDP keyword research, here's my suggestion:

Spend 20 minutes on your next book using the Amazon search bar method (free)
If you're serious about this, invest in Publisher Rocket ($100 one-time)
Track your results—keyword research without measurement is just guessing

For generating the books themselves, I use WriteAIBook.com. It handles the writing so I can focus on the publishing strategy—keywords, categories, covers, and ads. You can try it free with 30 credits, which is enough to generate 3 chapters and see if the quality works for you.

But the tool doesn't matter if you're publishing books nobody can find. Learn keyword research. It's the highest-leverage skill in KDP publishing.

Ready to Start Publishing Smarter?

Questions about KDP keyword research? I'm building a community of publishers who share what's actually working. Drop a comment or reach out—I read everything.

Try WriteAIBook.com Free

Before you read: blunt answers to common doubts

Is this saturated? Generic low-content books are saturated. Focused series in clear sub-niches still have room.

Does this still work? Yes, if you publish edited books consistently. One-off raw AI uploads usually fail.

Will I get banned? Not if you follow KDP policy: disclose AI usage, avoid spam, and label adult content correctly.

Is this a real business model? Yes. It is a workflow business, not a guaranteed-income promise.

How long until money? First sales can happen in weeks; stable income usually needs a catalog (often 20-50 books).

How much money realistically? Most consistent part-time publishers land in a few hundred to low four figures monthly after several months. Results vary by genre and execution quality.

Free tools

Steal the templates used by working KDP publishers

Outline fast → blurb fast → draft Chapter 1.

Free outline generator → Free blurb templates →

Generate Your First AI Book Today

Get 15 free credits (≈ 3 chapters). Test the output, then scale what works.

Try WriteAIBook Free →

No credit card required • Download DOCX • Unsubscribe anytime

Want a pen name that sells?

Get the free Pen Name PDF (double opt-in, unsubscribe anytime).

Get the free Pen Name PDF →