Which Amazon KDP niches still make money now?

Can Amazon KDP still make money now, or is it already too crowded?

Amazon KDP can still make money today, but the profitable part of the market has moved toward books people actively want to read, use or buy again.

The crowding is real. Bowker counted about 3.53 million self-published titles released in 2025, up 38.7% in a single year. Total U.S. print demand was nowhere near that pace. Circana BookScan recorded almost flat print sales in 2025, followed by another 0.3% decline in the first half of 2026.

That means the amount of self-published supply is exploding while the number of books people buy barely changes.

The good news is that demand underneath that flat headline is moving around a lot. Circana recorded adult fiction print sales up 5.7% in the first half of 2026. Games, activities and hobbies inside children's nonfiction grew 18.4%. The latest Association of American Publishers data tells a similar story from the revenue side: adult fiction was up 8.2% in the first half, children's and YA nonfiction rose 24.3%, and trade paperback revenue increased 8.4%.

Meanwhile, adult nonfiction revenue fell 3.5%, and Circana measured self-help print units down 24.3%.

So we are dealing with a market that is still producing clear winners, just very unevenly.

KDP model	What the market looks like now	Main reason
Series genre fiction	Strong	Readers can move through several books
Puzzle and activity books	Strong	Current demand is growing quickly
Children's educational activities	Strong	Screen-free formats are gaining sales
Narrow specialist nonfiction	Selective	Specific problems still create search demand
Generic journals and notebooks	Weak	Huge supply and almost no differentiation
Mass-produced generic books	Very weak	Production has become too easy to copy

Why is it so hard to tell which KDP niches actually make money?

Amazon does not publish KDP revenue by niche, so a bestseller page alone tells us very little about whether a niche is genuinely profitable for independent publishers.

This is where many KDP analyses go wrong.

Seeing several romance books high in Amazon's rankings confirms demand. It tells us nothing about the thousands of romance books earning almost nothing, the advertising required to reach those rankings, or how much money successful authors make from the rest of their catalog.

The same problem appears in low-content publishing. A notebook with a good Best Sellers Rank may be selling, but we cannot see how many almost identical notebooks the publisher tested before finding it.

We therefore have to combine several datasets.

Circana BookScan tells us where U.S. print units are moving. The Association of American Publishers gives us publisher revenue by category and format. Amazon publishes its current royalty rules and Kindle Unlimited fund. Bowker shows how quickly book supply is expanding. Written Word Media's latest large survey gives us a separate view into the economics of 1,346 independent authors.

That indie survey is particularly useful because the genres are unevenly distributed across income levels.

Romance made up 21% of respondents but around 44% of authors making more than $10,000 a month. Paranormal romance represented roughly 4% of respondents and around 11% of that highest-income group. Cozy mystery also appeared more frequently among authors earning at least $500 a month than among those earning almost nothing.

Those numbers have obvious selection bias: Written Word Media reaches serious commercial indie authors and the income figures are self-reported. We would never use them to estimate how much the average romance author earns.

They still help us identify which genres repeatedly show up among successful independent publishers.

Did Amazon make cheap KDP books less profitable?

Yes. Cheap KDP paperbacks have worse economics today, and the change hits commodity books particularly hard.

Amazon currently pays a 50% paperback royalty on Amazon.com when the list price is $9.98 or below. At $9.99 and above, the rate becomes 60%. Printing costs come out afterward in both cases.

Amazon's own example makes the effect easy to see. A 300-page black-and-white paperback priced at $9.98 earns roughly $0.39. Move the list price by one cent to $9.99 and the estimated royalty rises to $1.39.

That pricing threshold makes a big difference for products competing at the cheap end of Amazon.

A 333-page black-and-white paperback priced at $15 produces about $4 in royalty in Amazon's current example. An Expanded Distribution sale of the same book produces only about $1 because the royalty rate there is 40%.

KDP has also made the upper end of e-book pricing a little more flexible. Eligible e-books can now receive the 70% royalty rate at U.S. prices up to $12.99, rather than being forced into the lower rate once they went above the old $9.99 ceiling.

A useful $16 workbook, a $14 puzzle book or a novel that also earns Kindle Unlimited page-read revenue therefore has much more room to work with than a disposable $6.99 or $7.99 print product.

Example	Approximate current KDP royalty
300-page paperback at $9.98	$0.39
300-page paperback at $9.99	$1.39
333-page paperback at $15	$4.00
Same $15 paperback through Expanded Distribution	$1.00
Eligible e-book	Up to 70% of list price, after applicable delivery costs

Does Kindle Unlimited still make fiction one of the best KDP businesses?

Kindle Unlimited still gives serial fiction a huge advantage on KDP because one acquired reader can generate revenue across an entire catalog.

Amazon paid $71 million into the KDP Select Global Fund in July 2026 and $69.4 million in August. That is more than $140 million distributed in two months to participating authors based largely on pages read.

Those payments explain why book length and series depth matter so much in some KDP niches.

A reader who finishes one 400-page paranormal romance and immediately opens book two is more valuable than a customer who buys one isolated paperback and disappears. If the reader moves through six books, the economics improve again without requiring six separate customer-acquisition efforts.

Written Word Media's latest survey shows the same pattern from another angle. Around 80% of authors with only one to three published books earned less than $100 a month. Authors with at least 25 books reported median monthly income around $3,000, and more than 40% of them were making at least $5,000 a month.

Catalog size alone does not cause income. Successful writers survive long enough to build larger catalogs, so causality also runs the other way.

Even with that caveat, the survey found series especially common among the highest earners. That fits the mechanics of Kindle Unlimited very well.

Does romance still make serious money on Amazon KDP?

Romance is still one of the strongest KDP niches today, especially when an author writes a clear subgenre in a long series.

The broader book market remains healthy. U.S. print romance sales reached nearly 44 million copies in 2025, up 3.9%. Adult fiction then continued growing in 2026: Circana measured a 5.7% print increase in the first half, while the latest AAP data puts adult-fiction revenue growth at 8.2%.

Romance also keeps showing up where we would want it to show up in indie-specific data.

In Written Word Media's 2025 survey, romance represented 21% of all participating authors but approximately 35% of authors making $1,000–$5,000 a month, 38% of those making $5,000–$10,000, and 44% of authors above $10,000.

The pattern becomes even more interesting inside paranormal romance. Only around 4% of respondents wrote it, yet paranormal romance accounted for roughly 11% of authors above $10,000 a month. In the lowest income group, its share fell to about 3%.

I'd be much more confident in paranormal romance, sports romance, romantic suspense, dark romance or another established reader promise than in simply publishing a book described as “contemporary romance.”

Readers in these markets already understand tropes. Covers tell them what type of experience they are buying. Series can be consumed quickly. Kindle Unlimited suits the reading pattern.

Competition is brutal, of course. Romance authors have spent years professionalizing covers, blurbs, advertising, launch teams and rapid-release schedules. A weakly packaged book will look weak immediately.

Still, the evidence behind romance is stronger than the evidence behind most KDP niches.

Can cozy mystery still make money on KDP?

Cozy mystery remains one of the more convincing KDP niches because its readers are unusually compatible with a long-series publishing model.

The strongest evidence comes from indie authors themselves.

Cozy mystery accounted for only about 4% of respondents in Written Word Media's latest survey. Among authors earning more than $500 a month, however, its share rose to roughly 6–9%. Among authors making less than $100, it dropped to around 1.7%.

That pattern also appeared in the organization's previous annual survey, so we are seeing something more persistent than a one-year anomaly.

The genre's structure helps explain it.

A cozy mystery series can keep the same amateur detective, village, bakery, bookstore, pet or supporting cast for many installments. Readers know what they are coming back for, while each murder provides enough novelty to justify another book.

General mystery and thriller demand does not give us the same clean story. Suspense and thriller print sales weakened during 2025, showing why we should avoid using “mystery sells” as a blanket conclusion.

Cozy mystery stands out because the series structure encourages readers to keep going.

Are LitRPG, progression fantasy and science fiction good KDP niches now?

LitRPG and progression fantasy remain attractive KDP niches today, while broad fantasy requires more caution because recent print growth has been much more volatile.

Fantasy had an extraordinary 2024, with U.S. print units jumping 35.8%. Sales then fell 8.7% in 2025 as the blockbuster comparison became harder. That swing shows how misleading a single category-growth number can be.

Current speculative-fiction demand is more interesting below the headline.

Circana said science fiction, horror and manga-influenced fiction were performing well in early 2026. Science fiction had already grown 12.3% over the first nine months of 2025. Supernatural and horror was also the only young-adult fiction subcategory to grow in the first half of 2026, even as YA fiction overall fell 25.6%.

Then there is Dungeon Crawler Carl.

Matt Dinniman originally self-published the LitRPG series before Ace acquired print rights. One volume sold roughly 446,000 U.S. print copies in the first half of 2026, making it the seventh-bestselling print book in the country during that period.

That is obviously an outlier. What matters is what it reveals about the audience.

LitRPG readers are comfortable with very long books, explicit series progression and rapid continuation. Those characteristics also work beautifully with Kindle Unlimited's page-read model. Progression fantasy shares much of the same reader behavior.

That makes a tightly understood LitRPG or progression-fantasy audience much more interesting than the generic idea of “writing fantasy because fantasy is popular.”

Are puzzle books one of the best KDP niches right now?

Puzzle and logic books are currently among the strongest print-first KDP niches we found.

The recent numbers are unusually good.

In the first quarter of 2026, crafts, hobbies, antiques and games were one of only two adult nonfiction areas where Circana recorded growth. The category rose 9.6% while adult nonfiction overall was struggling.

The trend strengthened as the year progressed. Games, activities and hobbies inside children's nonfiction grew 18.4% in the first half. Circana has also singled out puzzles, logic and brain teasers as beneficiaries of the current move toward hands-on, screen-free entertainment.

This demand is showing up in real products.

Murdle turned murder-mystery logic puzzles into a recognizable publishing property and has sold hundreds of thousands of copies. The franchise then expanded into additional Murdle books and related formats such as Murdoku.

That tells us more than another generic Sudoku bestseller would.

A concept can give a puzzle book a reason to exist beyond the raw number of puzzles inside it. Murder mysteries can sit on top of deduction puzzles. Children's dinosaurs can sit on top of mazes. Travel can sit on top of word games. A strong visual world can connect an entire series.

KDP also treats puzzle and coloring books differently from low-content products such as blank journals. They contain substantive interior content and can build genuine intellectual property.

The crowded part remains the obvious part: “1,000 Sudoku Puzzles,” “Word Search for Seniors,” and hundreds of close variations.

Are children's activity books still growing fast enough to be worth publishing?

Children's activity books are one of the clearest current KDP opportunities because demand is growing much faster than the wider children's book market.

Circana recorded 16.7 million children's activity books sold in the U.S. in 2025, up 18%. Total children's book sales grew only about 3% over the same period.

The gap persisted into 2026. Through late May, activity-book units were running another 18% above the previous year.

The latest broader publishing data now reinforces that picture. Children's and YA nonfiction revenue rose 24.3% in the first half of 2026 according to the Association of American Publishers. Paperback revenue across the children's and YA segment was up 8.5%.

Publishers interviewed by Publishers Weekly attribute part of the activity-book growth to parents looking for portable, screen-free ways to keep children occupied and learning. Circana independently sees the same screen-free behavior across puzzles, coloring, guided journals and other hands-on formats.

For KDP, that creates several practical niches: handwriting, mazes, phonics, early mathematics, tracing, travel activities, scissor skills and tightly themed educational workbooks.

We should still separate those products from conventional picture books.

Written Word Media's indie survey found children's authors disproportionately concentrated in the lowest book-income tier and almost absent from the highest tiers. The sample skews toward commercial fiction authors, so we should treat that result carefully. Yet it matches the practical difficulty of indie picture books: illustration can be expensive, physical quality matters, and discovery often happens through schools, libraries, bookstores and parents rather than Amazon search alone.

An activity workbook has a much clearer Amazon query behind it. “First grade multiplication workbook” tells us exactly what the buyer needs.

Are adult coloring books and guided journals coming back?

Adult coloring, guided journals and other hands-on books are genuinely getting another wave of demand, although generic versions are still easy to drown in competition.

Circana's 2026 market research describes a broader return to analog activities. Adult coloring, journaling, needlework, crafting, puzzles and logic books are all benefiting as consumers look for ways to spend less leisure time on screens.

We would put much more weight on that cross-category pattern than on one coloring book suddenly ranking well on Amazon.

The trend is visible across adults and children, across several publishing categories, and across different types of tactile products.

Product quality decides how useful the trend is for KDP.

A blank “mindfulness journal” has thousands of substitutes. A guided journal for first-time fathers, grief after a particular type of loss, a 12-week running goal or another tightly defined situation can contain enough original structure to justify the purchase.

Coloring books face the same divide. Random collections of generic illustrations are easy to generate now. A consistent visual universe, recognizable humor, an unusual theme or a series gives buyers something harder to replace.

Printing economics also favor black-line coloring interiors because full-color KDP interiors cost much more to manufacture.

Is Christian publishing still a good KDP niche?

Christian devotionals, Bible-study books and related workbooks are currently supported by one of the strongest demand trends in nonfiction.

U.S. Bible sales reached roughly 19 million copies in 2025, their highest level in more than two decades. The broader group covering Bibles, devotionals and study guides increased from about 16.6 million units in 2019 to 30.8 million in 2025.

That is close to an 86% increase in six years.

The trend has survived into the current market. Religion was one of only two adult nonfiction areas with higher print units in the first quarter of 2026. The latest AAP results also show religious-publisher revenue up 1.6% in the first half after a 2.9% increase in 2025.

KDP publishers obviously face major specialists in the Bible market itself. Competing with established study Bibles or premium editions would make little sense for most new independents.

The surrounding niches are more accessible.

A 90-day devotional for a particular stage of life, a Bible-study workbook for a defined audience, faith-based children's activities, prayer resources or books written for a specific Christian community can target a much clearer reader.

Credibility matters more here than in a puzzle book. Readers are buying interpretation, guidance and trust.

Does self-help still make money on Amazon KDP?

Broad self-help is a difficult KDP niche right now, despite the enormous visibility of its biggest books.

The recent sales drop is striking. Circana recorded self-help print units down 26.3% in the first quarter of 2026 and 24.3% across the first half.

That decline needs context.

Self-help had a huge comparison problem after Mel Robbins' The Let Them Theory sold 962,291 print copies in the first quarter of 2025 alone. James Clear's Atomic Habits added another 253,341 copies during the same period.

The Let Them Theory was still the biggest nonfiction print book in the first half of 2026, with more than 566,000 copies sold. Atomic Habits added more than 252,000.

So readers have hardly abandoned self-help. A small number of giant authors simply capture an extraordinary amount of demand, while the category's recent growth was inflated by blockbuster sales.

That makes broad subjects such as productivity, confidence, habits and motivation uncomfortable places for a new KDP publisher.

We would rather go narrower until the title describes an identifiable problem. A guide for a newly promoted restaurant manager has clearer intent than another leadership book. A workbook for navigating one professional exam has clearer intent than another book about career success.

Useful credentials, lived experience, proprietary research or access to a specific community become increasingly valuable here because generic explanations are now extremely cheap to produce.

Can specialist nonfiction still make money when AI can answer almost anything?

Specialist nonfiction can still make money on KDP when the book gives readers something more dependable, organized or specific than an AI answer.

People rarely need a 250-page book to answer a simple factual question anymore. Search engines, YouTube, Reddit and AI can already handle a huge amount of generic “how to” information for free.

A specialist book can still earn its price when the problem is complicated enough that readers want a complete system rather than a collection of answers.

Think about a certification manual, a technical workbook, an operational guide for one type of small business, a detailed resource for a particular craft, or a practical book written around firsthand professional experience.

The economics can also be better than low-priced KDP products. A narrow professional book that comfortably sells for $17.99 or $24.99 has far more royalty room than a commodity paperback fighting to stay below $10.

Cookbooks fit the same logic. Broad air-fryer, Mediterranean-diet and beginner-recipe categories contain enormous competition. A chef, dietitian, creator or genuine regional expert can still build a compelling book around expertise or an existing audience.

AI has made the dividing line easier to see.

Amazon requires publishers to disclose AI-generated text, images or translations while allowing ordinary AI-assisted work such as brainstorming and editing. At the same time, Bowker recorded that 38.7% jump in self-published title output in 2025.

More books can now be produced faster, so information alone is becoming cheaper. Readers still pay for judgment, structure, trust and expertise.

Are generic notebooks, planners and AI-generated low-content books basically finished?

Generic low-content KDP is currently one of the weakest places for a new publisher to build a serious business, and AI is making that weakness worse.

Amazon classifies notebooks, planners, diaries, repetitive journals and logbooks as low-content books. These products are extraordinarily easy for competitors to reproduce, and many no longer even benefit from the economics that made the old mass-upload strategy attractive.

The royalty threshold we looked at earlier is one problem. Cheap paperback pricing now falls into Amazon's 50% tier at $9.98 and below in the U.S.

Supply is an even bigger one.

With roughly 3.53 million self-published titles appearing in 2025, there is little scarcity left in the ability to upload another gratitude journal, password tracker, lined notebook or weekly planner.

AI pushes that further. Covers, prompts, interior templates and basic text can now be produced in minutes. When production requires almost no scarce skill, competitors can follow any visible success very quickly.

There are still exceptions worth publishing. A maintenance log designed around a particular machine, a specialized occupational record book or a planner containing a genuinely useful system can solve a real problem.

At that point, buyers are paying for the system inside the book.

How much does having a KDP series actually change the economics?

A good KDP series can change the economics dramatically because the same reader has several opportunities to spend money or generate Kindle Unlimited pages.

Written Word Media's catalog data is hard to ignore. Around four out of five surveyed authors with only one to three books made less than $100 per month. At 25 books or more, median income climbed to roughly $3,000 a month, with more than 40% earning at least $5,000.

Again, we should avoid pretending that publishing 25 mediocre books automatically produces $3,000.

The useful part is the mechanism behind the correlation.

Suppose an author spends $4 in advertising or promotion to acquire a new reader and earns $4 from that first book. The first transaction barely works.

If half of those readers continue through several books, the original $4 acquisition suddenly supports much more revenue. No equivalent effect exists for a random notebook unless the publisher has built a recognizable brand.

Series also give marketing more leverage.

A promotion on book one can sell books two through eight. A mailing list built around one romance series can launch the next one. A successful children's puzzle concept can become progressively harder volumes. An early-learning workbook can move from preschool to kindergarten and first grade.

That makes repeat purchasing one of the most valuable traits a KDP niche can have.

Which Amazon KDP niches should we actually choose now?

The Amazon KDP niches with the best evidence today are series romance, cozy mystery, selected LitRPG and progression fantasy, puzzle books, children's activity books, distinctive analog activity books, faith-based resources for defined audiences, and narrow specialist nonfiction.

We have different levels of confidence inside that list.

Romance has the strongest indie-income evidence. It combines a large consumer market, current adult-fiction growth, heavy digital consumption and unusually high representation among successful self-published authors.

Cozy mystery has a smaller market but excellent series behavior and repeated over-representation in higher indie-income groups.

LitRPG and progression fantasy offer powerful Kindle Unlimited economics and extremely committed readers. The mainstream breakout of Dungeon Crawler Carl shows how far that audience can travel, although we should never treat one extraordinary series as an estimate of normal outcomes.

Puzzle books and children's activities are the strongest print-first opportunities we found. Their appeal comes from recent category growth rather than KDP folklore: children's activity sales increased 18% in 2025 and were still running 18% ahead in 2026, while the broader games, activities and hobbies segment posted similarly strong growth.

Faith publishing also has real demand behind it, with Bible-related consumption near multi-decade highs and religious publishing still growing while much of adult nonfiction struggles.

Specialist nonfiction is more fragmented. We would choose it when we have genuine expertise or a painfully specific customer problem, rather than because a keyword tool reports low competition.

At the bottom sit generic notebooks, interchangeable planners, broad self-help written without authority and mass-produced AI content. These formats have become much easier to supply without creating any corresponding increase in consumer demand.

KDP niche	Our confidence now	Best fit
Romance subgenres	Very high	Long e-book / Kindle Unlimited series
Cozy mystery	High	Recurring characters and long series
LitRPG / progression fantasy	High	Long books and Kindle Unlimited
Puzzle / logic books	Very high	Distinctive print series
Children's activity books	Very high	Educational or screen-free print products
Adult coloring / guided activities	High when differentiated	Recognizable concept or visual identity
Christian devotionals / study resources	High for credible authors	Specific faith audience
Specialist nonfiction	High when expertise is real	Expensive, specific problem
Specialist cookbooks	Selective	Expertise or existing audience
Broad self-help	Low for new entrants	Dominated by established authors
Generic planners / journals	Very low	Commodity product
Blank notebooks	Very low	Little reason to choose one over another
Generic AI-generated books	Very low	Competitors can reproduce them immediately

The biggest mistake would be searching for an untouched KDP niche. With millions of new self-published books appearing every year, almost every obvious niche already has competition.

We would search for valuable reader behavior instead.

Does a romance reader want seven books from the same author? Often, yes. Does a cozy-mystery fan want another case with the same characters? Very often. Will parents buy a new activity book when their child outgrows the previous one? Yes. Do puzzle fans want fresh puzzles once they finish the first volume? Clearly.

That repeat demand is now one of the best protections a KDP publisher can have.

Publishing itself has become extremely easy. Giving someone a reason to choose the next book from the same publisher is much harder.

That is where the money still is.