KDP Niche Research
How to find book ideas that hold up, before you write a single word.
Eight steps · An eight-signal scorecard · A thirty-minute screen
Most first books fail quietly, and they fail before a word is written.
The idea sounded good. Nobody checked whether readers were actually looking for it, whether the first page of results was already owned by established titles, or whether the promise could be kept. Three months later there is a finished manuscript and no readers, and the author concludes they are bad at writing when the real problem was chosen on day one.
Niche research is the work that prevents that, and it is not complicated. It is a sequence of observations, made in order, until the evidence points somewhere. This guide is that sequence. Work through it once and you will not again commit a season of your life to an idea you never tested.
What makes a KDP niche worth pursuing?
A promising niche sits at the intersection of four things. Miss any one of them and the opportunity usually collapses under real conditions.
Observable demand
Readers appear to be searching for and buying books on the subject.
Reachable competition
The market is not controlled entirely by famous authors, major publishers, or entrenched titles.
A visible gap
Readers repeatedly complain about something existing books omit, mishandle, or make unnecessarily hard.
A credible promise
You can move a defined reader from a recognizable starting point to a useful outcome, without exaggeration.
Low competition alone is not enough. Sometimes competition is low because demand is low. And high demand is not automatically attractive if every visible result has a powerful brand, thousands of reviews, and a product you cannot improve upon.
What is a KDP niche?
A topic is broad. A niche is a focused market inside that topic. A strong niche usually combines a reader, a problem or desire, a context, and a result.
Broad topic + specific reader + specific problem or desire + context or constraint + useful outcome
You do not need every element in the final title, but using the formula during research prevents you from mistaking a large category for a workable publishing opportunity.
The eight-step niche research process
Run these in order. Each step narrows the field and builds a small body of evidence you will use to score the opportunity at the end.
Start with a market you can serve
Begin with two or three broad subject areas. Do not choose only from a list of allegedly profitable niches; those lists are snapshots, and public opportunities become crowded quickly. Start where at least one advantage is true:
- You have direct experience, credentials, a distinctive framework, or access to credible experts.
- You understand the reader because you have faced the same problem.
- You can research the topic responsibly and support consequential claims with trustworthy sources.
- You can see several related books or follow-up titles that could serve the same audience.
Niche down into reader-and-outcome combinations
Generate at least ten narrower combinations from each broad topic. Avoid changing only one adjective; change the audience, use case, constraint, experience level, or desired transformation. A useful prompt: what specific type of person wants this result, under what circumstances, and what makes the usual advice unsuitable for them?
Define the reader transformation before checking keywords
Write the journey in two sentences. Point A: what is true about the reader before the book, including their uncertainty or limitation. Point B: what they can understand, decide, create, or do after. Then one promise sentence. This promise becomes your validation filter.
Collect several independent demand signals
No single Amazon number proves demand. Build a small body of evidence, and record the date, marketplace, search phrase, and format for every observation.
- Amazon search results for the exact phrase and close variations.
- Autocomplete language, treated as vocabulary research, not proof of sales.
- Best Sellers Rank compared across more than one observation.
- Review recency across several relevant books.
- Several distinct selling books, not one author-brand effect.
- External signals: Google Trends, forums, communities, podcasts, newsletters.
Measure the competition you actually need to beat
Do not use total search results as your only metric. Examine the first page of highly relevant results: how many books serve the same reader and promise the same outcome, how strong the title, cover, description and reviews are, whether independent publishers are visible, and whether you can create a meaningfully better product. Your real competition is the set of books a shopper would reasonably compare with yours.
Mine reader reviews for unmet needs
Negative and mixed reviews are often more useful than bestseller lists. Read a representative set of one-, two-, and three-star reviews across several close competitors and tag recurring complaints such as too basic or padded, missing steps or templates, outdated facts, over-promising titles, or poor organization. Look for repeated patterns, not one angry review.
Test the economics and production reality
A niche may have demand but still be unattractive for your model. Estimate product shape (format, price range, page count, trim size, color or images), cost structure (research, editing, design, permissions, printing, advertising, your time), and durability (how often content needs updates). Do not confuse a high list price with high profit.
Validate the title, metadata, and compliance risks
A clear title tells the reader what the book is about; clarity usually beats cleverness at thumbnail size. Search for confusingly similar titles and covers, screen for trademark confusion, choose only relevant keywords (KDP allows up to seven) and categories, flag higher-risk subjects for stronger sourcing, and plan for KDP's AI-generated content disclosure where required.
The Galley niche validation scorecard
Score each signal from 0 to 2. The score is a decision aid, not a revenue prediction. Write one sentence of evidence beside every score in your working notes.
Demand evidence
Little or no observable activity
Some signals, but inconsistent
Several relevant books and external signals show activity
Competition
Entrenched or dominated
Competitive but potentially reachable
Visible room for a differentiated entrant
Reader clarity
Broad or undefined audience
Audience partly defined
Specific reader and context are clear
Problem urgency
Nice-to-know topic
Useful but not pressing
Meaningful problem or desired outcome
Market gap
No clear unmet need
Possible gap with limited evidence
Repeated complaints reveal a clear opening
Promise feasibility
Too broad, risky, or unrealistic
Deliverable with refinement
Focused promise you can credibly fulfill
Economic fit
Cost or pricing looks unfavorable
Economics require caution
Format, price, and production needs appear workable
Publisher fit
No authority, access, or advantage
Researchable with added support
Relevant expertise, experience, or access
13 to 16 indicates a strong candidate for deeper validation. 9 to 12 suggests caution and refinement. 0 to 8 usually means the idea needs a different audience, promise, or market. A serious compliance or rights problem can override the number.
Take the scorecard with you.
This page has the method. The designed field guide has it in the form you actually use: sixteen pages built to be printed, kept beside you while you research, and filled in with a pen. Same content, better working copy.
A worked niche-down example
Travel
Digital nomad guide. Audience and outcome still too broad, and facts vary by country and date.
A 30-day pre-departure planning guide for first-time remote workers preparing to live abroad.
The reader wants to try location-independent work but is overwhelmed by preparation, budgeting, documents, work routines, and destination trade-offs.
The reader leaves with an organized departure plan, a realistic budget framework, a document checklist, and criteria for comparing destinations.
Possible differentiation. Decision worksheets, a dated source register for changeable information, budget scenarios, and a country-comparison framework instead of generic lifestyle inspiration. The narrower concept is easier to outline, research, title, and validate, and it exposes the evidence burden: visa, tax, health, insurance, and residency information must be sourced carefully and reviewed for freshness.
Common niche research mistakes
Choosing a niche from passion alone
Interest helps you write a better book, but it does not establish demand. Combine enthusiasm with observable market evidence.
Treating low competition as the goal
The goal is reachable competition alongside genuine demand. An empty market may be an opportunity, or a warning.
Copying arbitrary BSR or review thresholds
Rank is dynamic, category- and marketplace-dependent, and relative. Use it as one comparative signal collected over time.
Counting every search result as a competitor
Measure close substitutes: books serving the same reader, problem, format, and promise.
Mining keywords but ignoring readers
A keyword attracts a click, but the book must satisfy a person. Reviews and samples reveal the product you need to build.
Writing the book before defining its promise
Without a clear Point A and Point B, chapters become a collection of related information rather than a transformation.
Using AI output as evidence
AI can organize observations and suggest sub-niches. It cannot prove demand. Treat generated claims as hypotheses until supported.
Chasing today's most profitable niches list
Public lists are copied quickly and rarely account for your capabilities, format, or costs. Learn the method instead.
A 30-minute first-pass validation sprint
This quick screen is not the final decision. It tells you whether an idea deserves deeper research. Work top to bottom.
- Write the specific reader, Point A, Point B, and a one-sentence promise.
- Search the main phrase and three close variations in the target Amazon marketplace.
- Identify the ten most relevant competing books, not merely the first ten results.
- Record format, price, publication date, review count and recency, rank signals, title promise, and publisher type.
- Read negative and mixed reviews across at least three close competitors and tag repeated complaints.
- Check whether you can describe a meaningful advantage in one sentence.
- Run the scorecard and choose PASS, CAUTION, or NO-GO for further investment.
If the result is caution, do not immediately discard the broad market. Try changing the reader, outcome, format, context, or level of specialization. Often the opportunity is one layer deeper.
From niche verdict to book blueprint
Once a niche passes validation, convert the evidence into a book plan. The research should shape the manuscript, not sit in a forgotten spreadsheet. Before committing, tick off each of these.
- Turn the reader's Point A and Point B into the book's central promise.
- Translate recurring review complaints into chapter outcomes or quality requirements.
- Give each chapter one job and remove overlapping sections.
- Create an evidence plan for factual and changeable claims before drafting.
- Write the description and subtitle only after checking the outline can deliver every promised benefit.
- Preserve the language readers use, but avoid stuffing metadata or manuscript text with repetitive keywords.
You have the method. The method takes an afternoon.
That is the honest cost of doing this by hand, and it is worth paying when you are choosing between two or three ideas. When you are choosing between twenty, the arithmetic changes.
Galley runs the same page-one checks and returns a verdict in about a minute. ENTER, CAUTION, or AVOID, with the numbers behind it, so you can see exactly why.
Free. No card.
Frequently asked questions
- What is KDP niche research?
- The process of evaluating a specific Amazon book market before publishing. It examines reader demand, relevant competition, pricing and format, unmet needs, keywords, production requirements, and your ability to deliver a differentiated book.
- How do I find profitable KDP niches?
- Start with a broad topic, narrow it to a specific reader and outcome, collect several demand signals, analyze close competitors, mine repeated review complaints, test the economics, and score the evidence. Profitability is never guaranteed, but the process reduces intuition-only decisions.
- Is a low-competition KDP niche always better?
- No. Low competition can mean an underserved market, but it can also mean weak demand. Look for the combination of observable demand, reachable competition, and a gap you can serve meaningfully.
- Can I do Amazon KDP niche research for free?
- Yes. Amazon search, book detail pages, samples, customer reviews, category pages, Google Trends, community discussions, and the USPTO trademark search support a solid manual workflow. Paid tools may speed up collection but do not replace judgment.
- Should I use Best Sellers Rank to estimate demand?
- Use BSR as one relative signal, not a precise forecast. Amazon states that rankings reflect recent and historical customer activity, weight recent activity more heavily, and can change even when a book's activity stays the same.
- How many KDP keywords can I add?
- KDP currently allows up to seven keyword entries. They should accurately relate to the book, and Amazon advises publishers to avoid vague, irrelevant, or misleading keywords.
- Can AI help with KDP niche research?
- AI can help brainstorm sub-niches, organize competitor observations, cluster review complaints, and draft hypotheses. Verify every market and factual claim independently. KDP currently requires disclosure of AI-generated text, images, or translations.
- How narrow should my book niche be?
- Narrow enough that a reader immediately recognizes the book as relevant, but broad enough that observable demand and a complete, useful book still exist. If you cannot describe the reader and promised outcome in one sentence, the niche is probably still too broad.
The best time to discover that a niche is weak is before you write 30,000 words.
A clear verdict gives you permission to proceed, refine the angle, or walk away while the cost is still low.
- Amazon KDP Content Guidelines
AI-generated content disclosure, rights, quality, and customer-experience guidance.
- Amazon KDP Metadata Guidelines
Title, description, keyword, category, cover, and ISBN guidance.
- Amazon KDP keyword guidance
Current keyword-entry guidance.
- Amazon KDP Sales Ranking
How Best Sellers Rank and category ranks behave.
- USPTO Trademark Search
Federal trademark search system and search guidance.
Recheck all Amazon policy references at publication and on a schedule, because KDP rules and interfaces can change.