How to Vet a KDP Niche Using Only Free Amazon Data

May 26, 2026

You’ve seen the advice a hundred times. Find a keyword with low competition and you’re set. I went along with it for a while, until I started pulling the actual numbers instead of trusting my gut. It mostly doesn’t hold up. Low competition, more often than not, just means nobody is buying.

Here’s the check I run now before I commit to a niche. Everything in it is free and already sitting on the Amazon page. No paid tool, no invented metrics.

The three things I check

There are three, and you need all three. Any one of them on its own will lie to you.

First, autocomplete, for demand. Start typing your idea into the Amazon search box and watch what it fills in. Those suggestions are real things people search for. If Amazon won’t even complete your phrase, that is usually a sign nobody is looking for it.

Second, the result count, for competition. Search the term and read the “1-16 of over X results” line near the top. That is roughly the size of the field you’d be up against.

Third, the top two or three books themselves. Their review counts tell you whether you could make something better. Their Best Sellers Rank tells you whether they’re actually selling or just sitting there.

One honest note before the numbers

Amazon doesn’t publish search volume. So when a tool shows you “1,300 searches a month” for a keyword, it is guessing and dressing the guess up as a fact. The three things above are different, because you can check them yourself. That is the line I draw. I trust what I can verify and treat the rest as the estimate it is.

Broad keywords are a bloodbath

And bolting a long-tail onto the end usually doesn’t rescue you. Here are real counts I pulled on May 26, 2026, all paperback:

Tacking a demographic onto a giant keyword tends to do almost nothing. “prayer journal for teen girls” is still north of 30,000 books. You’re swimming in the same deep end.

The kind of qualifier that actually works

What does thin the field is a word that changes the book itself, not just the title.

Waterproof isn’t a marketing word, it’s a different product. Most of those 40,000 books aren’t really competing with a waterproof one. The field drops from forty thousand to about a thousand.

Where low competition fooled me

Here’s the one that taught me the lesson. Sobriety journals looked perfect. Under 800 competing books, and the top result had all of 6 reviews. Wide open, or so I thought.

Then I looked at the sales ranks.

A Books rank somewhere between 82,000 and over a million isn’t a money maker. It’s the occasional sale. Rank isn’t a copy count, so I won’t pretend it converts to exact sales, but this far down the list the leaders in a niche are barely moving. The competition was thin because the demand was thin.

So my honest take on sobriety journals is a maybe, not a yes. You could win it, the top books are weak, but the whole thing is small. It’s only worth your time if the book is cheap to make and you’re stacking a dozen little niches like it, not betting the farm on one.

What I’d go back and tell myself

Stop trusting any single number on its own. The competition count fooled me plenty. Review count did too. Rank fooled me the other way, looking weak when the niche was just empty. You only get a real read when you hold all three up together.

And the big one. Low competition is often a polite way of saying nobody is buying, and sales rank is the tiebreaker that tells you which kind you’re looking at. Autocomplete is the one free demand signal Amazon actually hands you, so use it, and stay honest with yourself that anything fancier about search volume is a guess.

I ended up turning this into a thing, mostly because I got tired of running the same check across a dozen open tabs. It’s a browser overlay that drops these numbers right onto the Amazon page and, on principle, won’t make up a search volume it can’t stand behind. It’s called QuillRadar, and it isn’t live yet. If you want a look when it is, the waitlist is here.

Numbers pulled from Amazon.com (US, paperback) on May 26, 2026. Competition counts and sales ranks drift, so treat them as a snapshot and check again before you commit.