Amazon Publishing KDP: Niche Selection & Market Research

Okay so here’s the deal with niche selection for KDP – most people overthink this and end up either picking something way too competitive or something so obscure that like three people on earth would buy it.

I’m gonna walk you through exactly what I do when I’m hunting for a new niche, and honestly I just did this last week because I wanted to test a theory about Q1 niches.

Start With Amazon’s Search Bar Because It’s Free Data

First thing – open Amazon in an incognito window. You want clean search results without your browsing history messing things up. Type something broad like “journal for” and just let the autocomplete do its thing. Amazon’s showing you what people are actually searching for right now.

I keep a Google Doc open and just dump everything interesting I see. Don’t filter yourself yet. I saw “journal for anxiety” pop up and my first thought was nope, too competitive. But then I kept digging and found “journal for anxiety and overthinking” which is more specific, and then “journal for anxiety for teens” which is even better.

The autocomplete suggestions are literal gold because Amazon only shows you searches that happen frequently enough to matter. If it’s appearing there, people are looking for it.

Amazon Best Sellers Rank Is Your Friend

So BSR – this number tells you how well a book is selling relative to other books in that category. Lower number = more sales. A book ranked #1 in a category sells way more than #100,000 overall.

Here’s what I look for: books ranked between 10,000 and 100,000 overall. That sweet spot means they’re selling consistently but there’s still room for new products. If the top 20 books in a niche are all under 5,000 BSR, that’s actually kinda intimidating because you’re competing with established sellers who probably have reviews and brand recognition.

Wait I forgot to mention – you need to check multiple books in the niche, not just the top one. Sometimes there’s like one mega-successful book and then everything else is ranked 500,000+. That’s a red flag. You want to see several books doing well, which tells you there’s actual sustained demand.

The Calculator I Actually Use

I use this free Chrome extension called DS Amazon Quick View. Shows me the BSR right on the search results page without clicking into each listing. Saves so much time it’s ridiculous.

For estimating sales from BSR, there’s rough formulas floating around but honestly they vary. Generally speaking, a book ranked around 50,000 might sell 10-15 copies a day. At 100,000 maybe 5-7 copies. These aren’t exact but they give you a ballpark.

I multiply estimated daily sales by my royalty (usually $2-4 for a low content book) and see if the monthly revenue makes sense. If the #10 book in a niche is only making like $100/month, I’m probably passing unless I can create something genuinely better.

Checking Competition Quality Not Just Quantity

This is gonna sound weird but I actually want to see some competition. No competition usually means no demand. What I’m looking for is BAD competition that I can beat.

Open the top 10-20 books in your potential niche. Look at:

• Cover design – is it professional or does it look like someone made it in Word?
• Page count – are they skimping with 50 pages when they could offer 120?
• Interior preview – Amazon lets you peek inside, so check if the interior is actually good
• Book description – is it keyword-stuffed garbage or actually written for humans?
• Price point – are they all priced at $5.99 when you could offer better value at $7.99?

If you see books with terrible covers ranking well, that’s opportunity. Means the demand is strong enough that even mediocre products sell.

I found this niche last month – gratitude journals for nurses. The top books had covers that looked like stock photos slapped on with text. Interior was basic lined pages. But they were ranked around 40,000 BSR. I made something with better design, added some nurse-specific prompts and quote pages, and it’s been selling 3-5 copies daily since launch.

Keyword Research Without Spending Money

Okay so there’s paid tools like Publisher Rocket which I do use sometimes, but you can do a lot for free.

Amazon’s search bar again – type your main keyword and add letters. Like “gratitude journal a” then “gratitude journal b” and so on. You’ll find variations you didn’t think of. “gratitude journal bundle” “gratitude journal blue” “gratitude journal boys” – these all tell you what modifiers people use.

Also check the “customers also searched for” section on product pages. And the “customers also bought” section. Both show you related niches and keywords.

I keep a spreadsheet with three columns: keyword, search volume estimate (just my gut feeling of high/medium/low based on autocomplete), and competition level. Then I’m looking for high search volume + medium or low competition.

The 7 Backend Keyword Slots

You get 7 keyword boxes on KDP when you upload. Don’t waste these on single words. Use phrases people actually search. “Gratitude journal for women daily” is way better than just “gratitude” in one box and “journal” in another.

My cat just jumped on my keyboard sorry – anyway, I use variations and long-tail keywords in those boxes. Think about how people search. They don’t type “journal” they type “journal for men with prompts lined pages” or whatever.

Seasonal vs Evergreen Niches

This is important. Some niches spike hard during certain times and die otherwise. Anything teacher-related blows up July-September. Wedding planners spike January-April. Christmas stuff is obvious.

I do both but my portfolio is like 70% evergreen 30% seasonal. Evergreen stuff like habit trackers, password logs, gratitude journals – people buy these year-round. The cash flow is predictable.

Seasonal can make you a ton of money in a short window but then it’s crickets. I have Halloween activity books that make $800-1000 in September-October and maybe $50 total the rest of the year.

To check if something’s seasonal, scroll through the review dates on top books. If a book has 50 reviews and 45 of them are from November-December, you know it’s seasonal.

Testing Your Niche Theory Before Going All In

I never create like 10 books in a niche before validating it. Here’s what I do:

Make ONE book. Price it competitively. Give it a solid cover and interior. Run a small AMS ad campaign if you want faster data – like $5/day for a week. See what happens.

If it gets no impressions on your ads, the search volume might not be there. If it gets impressions but no clicks, your cover probably sucks. If it gets clicks but no sales, could be your price, your preview, or just weak demand.

I tested a niche for budgeting journals targeted at college students last year. Made one book, ran ads, got decent impressions and clicks but conversion was terrible. Turns out college students don’t really buy physical budgeting journals – they use apps. Learned that for $30 in ad spend instead of creating 5 books and wasting a week.

The Three-Book Approach

If that first book shows promise, I create two more variations. Different cover styles or slightly different angles on the same niche. Like if my first gratitude journal was floral and feminine, I’d make one that’s minimalist and gender-neutral, and one that’s more masculine.

This helps you capture different customer preferences within the same niche. Plus Amazon’s algorithm seems to like when you have multiple books – they’ll show your other books in the “more from this author” section.

Red Flags That Make Me Skip a Niche

• Top books all have 500+ reviews – you’re not catching up anytime soon
• First page of results dominated by major publishers or celebrity authors
• BSR of top 20 books is all over 200,000 – demand is too weak
• Every book looks identical – means there’s a template everyone’s using and the market is saturated
• Trademark issues – anything Disney, sports teams, brand names, you’re risking takedowns

Oh and another thing – if you need specialized knowledge to create the content properly, that’s usually a skip for me. Like medical journals for specific conditions, legal planners, financial tracking with complex formulas. Not worth the liability risk.

Where I Find Niche Ideas Outside Amazon

Etsy is underrated for this. Search planners, journals, logbooks on Etsy and sort by best selling. These sellers have already validated demand. You can’t copy their designs obviously, but you can see what niches are working.

Pinterest trends show you what people are interested in. Reddit communities are gold – lurk in subreddits related to hobbies, professions, life situations. When you see people asking “does anyone have a good journal for X” that’s a niche opportunity.

I found the nurse journal thing from a nursing subreddit where someone asked for planner recommendations and got like 50 comments. Clearly there was demand.

Facebook groups too. Parenting groups, hobby groups, professional groups. Just observe what people need or complain about not having.

Validating Demand With Google Trends

Before I commit to a niche, I’ll check Google Trends. Type in your main keyword and see if interest is growing, stable, or declining. You want stable or growing.

Also check if it’s US-focused or global. Most of my sales come from the US so I optimize for that market, but UK and other English-speaking countries can add nice supplementary income.

The regional interest map is helpful too. If something’s only popular in like Montana and nowhere else, the market might be too small. You want broad geographic interest.

Pricing Research Is Part of Market Research

Look at what the top books charge. Most low content books are between $5.99-$12.99. Your cost to print through KDP depends on page count and color vs black and white.

I use KDP’s royalty calculator to figure out my profit at different price points. Generally I aim for at least $2-3 profit per sale after printing costs. Sometimes more if it’s a premium product.

If everyone in the niche is priced at $6.99 and I can make a better product, I might price at $8.99. The extra $2 filters for customers who value quality. Or sometimes I’ll match the average price but offer more pages or better design – perceived value is huge.

The Actual Process I Follow Every Time

Since this is kinda scattered, here’s my step-by-step:

1. Brainstorm broad niche ideas using autocomplete and bestseller lists
2. Check BSR of top 20 books in each niche – looking for that 10k-100k sweet spot
3. Analyze competition quality – can I do better?
4. Gather keywords from Amazon, check seasonality
5. Validate with Google Trends and outside platforms
6. Calculate estimated revenue vs effort required
7. Create one test book if everything looks good
8. Evaluate results after 2-4 weeks
9. Scale with variations if it works, abandon if it doesn’t

I spend maybe 2-3 hours on this research before creating anything. Used to skip this step when I started and wasted so much time on books that never sold. Now I’d rather spend 3 hours researching than 10 hours creating something nobody wants.

One last thing – keep a swipe file of niches that almost worked or that you want to revisit. I have a note on my phone with like 40 niche ideas. Some I tested and they weren’t quite right but might work with a different angle. Some I haven’t gotten to yet. It’s helpful when you’re stuck for ideas or when you need something to test during a slow month.

The whole game is really just finding gaps between what people want and what currently exists. When you spot those gaps, that’s where you make money.