Amazon KDP Keywords 2026: 7 Slots, Long-Tail + Genre Lists

amazon kdp keywords kdp keywords kdp keyword research
Amazon KDP Keywords 2026: 7 Slots, Long-Tail + Genre Lists

Amazon KDP keywords are seven optional phrases, each up to 50 characters, that you enter in the KDP dashboard. They tell Amazon's 2026 search stack — A9 plus the COSMO semantic layer and Rufus conversational AI — which reader queries should surface your book. Don't repeat words already in your title, subtitle, or categories.

This guide is the practical version of Amazon's official keyword help page, with the parts the official page leaves out: how to allocate the 7 slots, genre-specific starter lists, an honest tool comparison, and the post-launch rescue workflow nobody else publishes.

What KDP keywords are (and what changed in 2026)

The mechanics: 7 text boxes, 50 characters each, indexed but not displayed to readers. The 7 slots haven't changed in years, but in 2024–2026 Amazon layered two new systems on top of A9 that change what works inside them:

  • COSMO — a semantic / knowledge-graph layer launched in 2024. It maps related concepts so a search for cozy mystery cat can surface a book whose keywords say amateur sleuth bakery owner with feline companion even without exact-word overlap.
  • Rufus — a conversational AI shopping assistant rolled out late 2024 and expanded through 2026. Rufus answers natural-language queries like "what's a good fantasy series for fans of Brandon Sanderson?" by combining your keywords, description, reviews, and category signals.

The tactical implication: single-word stuffing is dead. The phrases that work in 2026 are the ones a real reader would type or speak — descriptive, specific, multi-concept.

How to add or update keywords in your KDP dashboard

The mechanic is simple, but the timing matters.

  1. KDP Bookshelf → find your book → "..." → Edit eBook Details (or Edit Paperback Details / Edit Hardcover Details — keywords are set per format).
  2. Scroll to Keywords. Seven text fields, labeled 1–7.
  3. Type or paste your phrase. The field accepts up to 50 characters including spaces.
  4. Save and publish. For backend changes, you don't need to upload a new manuscript — Amazon re-indexes the existing one.

Timing notes most authors miss:

  • Updates typically index within 24–72 hours, occasionally longer for newly published books still in initial review.
  • After any change, wait 5–7 days before judging impact (per Kindlepreneur's documented testing). Sales rank moves on weekly cycles; same-day "did it work?" panic is wasted attention.
  • Don't change all 7 at once. If you do, you can't attribute results to any single change. Two or three per cycle is the maximum that's still measurable.

The 7-slot allocation framework

Treat the 7 slots as a portfolio. Three roles, not seven copies of the same idea:

Slot role How many What it captures Example (cozy mystery)
Subgenre/trope phrases 2–3 Reader-vocabulary descriptions of what your book actually is cozy cat mystery small town, amateur sleuth bakery series
Comp / "books like" phrases 2–3 The way readers search for adjacent authors or series books like agatha raisin, cozy series with recurring characters
Niche-modifier phrases 1–2 Highly specific qualifiers your top comps don't cover clean cozy mystery no profanity, cozy mystery female protagonist over 50

A common rookie pattern is filling all 7 with subgenre variations of the same idea (cozy mystery, cozy mystery series, cozy mysteries for women, best cozy mystery...). Amazon's semantic layer already groups those — you're using 7 slots to send 1 signal. Spread across the three roles instead.

Two hard rules from Amazon's keyword policy:

  • Don't repeat words from your title, contributors, or categories. Amazon already indexes those. Use the slots for phrases you couldn't fit on the cover.
  • Don't make subjective claims. Words like bestseller, best, new, or award-winning are explicitly prohibited.

What about your series name?

If your book is in a series, Amazon already indexes the series name from the Series field during KDP setup. Burning a keyword slot on Sweetbriar Cove book 3 or Detective Yang series is wasted — readers searching the series name find your book through that field, not your keywords.

The slot is better spent on the flavor of the series readers search for. Instead of the series name, use phrases that capture what makes the series distinctive: small town series with recurring cat character, multi-generational family saga rural, procedural detective series female lead Pacific Northwest. Those phrases ride on COSMO's semantic layer to surface your series for readers who don't know it by name yet.

One legitimate exception: if your book has a shared-universe crossover with another series readers already search for, a slot like crossover with [other-series-flavor] can pull discovery traffic. Otherwise: keep the series name in the series field, keep your 7 slots free for discovery phrases.

How to find long-tail Amazon KDP keywords (autocomplete + reverse-ASIN)

The single most useful free research tool is Amazon's search box. The "type a phrase, watch the dropdown" trick is older than KDP itself and still beats most paid tools for ideation.

Amazon Kindle Store search bar showing autocomplete suggestions for sci-fi keywords

The method:

  1. Switch the search dropdown to Kindle Store (not "All Departments") so the suggestions are book-weighted, not blender-weighted.
  2. Type a 2–3 word phrase from your starter list — e.g., enemies to lovers.
  3. Watch the dropdown. Amazon shows the most common longer queries that begin with what you typed: enemies to lovers romance, enemies to lovers fantasy, enemies to lovers workplace, enemies to lovers hockey.
  4. Type the next letter of the alphabet after a space — enemies to lovers a, then b, then c — to see which extended phrases Amazon will autocomplete. (This is the "alphabet soup" method indie authors have used since 2015.)
  5. Copy promising phrases into a scratch list. Don't filter yet — quantity first.

Reverse-ASIN as a complement. Find 3–5 bestselling books in your niche (BSR under 100,000 in their main category is a good filter). Read their full Amazon listing — title, subtitle, description, A+ Content, and the categories they're shelved under. The phrases that recur across multiple top-sellers are almost certainly some of the keywords driving their traffic. Add them to your list.

Use incognito mode. If you're logged into your normal Amazon account, autocomplete is personalized to your recent browsing. Open an incognito tab to see the suggestions a stranger would see. Your "logged-out" view is what Amazon shows your potential reader.

Subscribe & Get Your Free Marketing Plan Template

Receive regular updates on marketing best-practices, AI shortcuts, and get our proven 4-phase marketing roadmap for free.

Unsubscribe anytime.

Genre-specific keyword starter lists

Most guides give the same two examples (historical romance, fantasy novel) and stop. That's not enough to actually fill 7 slots. Below: candidate phrases by genre, written in the language readers actually use on amazon.com. Treat these as starting points — always validate with Amazon autocomplete before using.

Romance

enemies to lovers contemporary romance
second chance romance over 40
billionaire romance with HEA
small town romance series
paranormal romance shapeshifter alpha
workplace romance slow burn
sports romance hockey enemies to lovers
dark romance morally grey hero
forbidden romance age gap
clean wholesome romance no spice

Fantasy & sci-fi

epic fantasy with dragons and magic
military science fiction space opera
dystopian thriller medical theme
sword and sorcery quest standalone
LitRPG progression fantasy first book
cozy fantasy bookshop tea magic
urban fantasy detective series female lead
post apocalyptic survival fiction
cyberpunk noir AI antihero
high fantasy political intrigue series

Cozy mystery

cozy cat mystery small town series
amateur sleuth bookshop owner cozy
culinary cozy mystery recipes included
knitting mystery cozy small town
dog mystery series amateur sleuth
holiday cozy mystery christmas village
small town witch cozy paranormal
tea shop cozy mystery british village
bakery cozy mystery recurring characters
clean cozy mystery no profanity

Thriller & suspense

psychological thriller unreliable narrator
domestic suspense husband secrets
police procedural detective female lead
legal thriller courtroom drama
historical mystery 1920s detective
medical thriller hospital conspiracy
nordic noir detective series translated
small town suspense missing person
spy thriller cold war female protagonist
slow burn psychological suspense

Self-help & personal growth

emotional healing after divorce workbook
setting healthy boundaries relationships
anxiety workbook for women
mindfulness meditation beginners daily
journaling prompts shadow work
codependency recovery toolkit
self love practical exercises women
trauma healing somatic exercises
midlife reinvention career change women
boundaries with difficult family members

Self-help keywords lean heavily on emotional outcomes — boundaries, relationships, emotional healing, trauma recovery, codependency — combined with format qualifiers (workbook, journal, exercises, daily practice) and audience descriptors (women over 40, for men, beginners). The phrase patterns that win in this genre fuse the pain point with the modality. A reader searching setting healthy boundaries relationships is buying a practical resource; a reader searching emotional healing alone is browsing. Aim slots at the practical-resource end of the spectrum.

Nonfiction how-to

passive income online business beginners
productivity system for entrepreneurs ADHD
public speaking confidence introverts
time management working parents
financial independence early retirement
home organization minimalist living
gardening for beginners small space
sourdough baking beginner step by step
beginner watercolor techniques flowers
parenting strong willed toddlers

Children's, middle grade, and YA

middle grade fantasy adventure series
chapter book early reader animals
picture book bedtime story diverse
YA contemporary romance enemies to lovers
middle grade graphic novel friendship
YA dystopian survival series strong heroine
early reader phonics decodable
picture book preschool emotions feelings
middle grade mystery sleuth siblings
YA fantasy magic school enemies to lovers

A starter list isn't a finished list. Validate each candidate with the autocomplete + reverse-ASIN method above before committing it to a slot.

Free vs paid KDP keyword tools

There's a real debate in the indie community about AI-generated keywords versus data-driven tools. Dave Chesson at Kindlepreneur is openly skeptical of using general-purpose ChatGPT for keywords — his point is sound: ChatGPT has no Amazon search-volume data, so when it suggests "trending" phrases, it's pattern-matching, not measuring.

That's true for general chatbots. The honest picture for 2026:

Tool Type Price (mid-2026) What it's good for What it can't do
Amazon autocomplete Free, manual $0 Real reader phrasing, instant validation No volume numbers, manual labor
Our KDP Keyword Generator AI, free $0 30-second starting list from your synopsis No Amazon search-volume data
Publisher Rocket Data tool $199 one-time Actual search volume + competition + AMS keyword research Doesn't write phrases for you
KDSpy Data tool, browser extension $79 one-time Quick competitor BSR + category snooping Lighter on keyword volume than Rocket
Helium 10 (Cerebro) Data suite $39/mo add-on or $99+/mo full suite Reverse-ASIN at scale, Amazon Ads keyword work Overkill for a single book
BookBeam Data suite ~$348–576/yr Niche analysis + tracking Pricing rose meaningfully in 2025
K-lytics Genre reports $37–97/mo Genre-trend reports curated by Alex Newton Not a per-book keyword tool

The honest workflow:

  1. AI generator for ideation (free, ~30 seconds). A focused tool like our generator reads your synopsis and proposes 7 candidate phrases — useful as a starting point when you're staring at empty boxes.
  2. Amazon autocomplete for validation (free, instant). Run each AI-generated phrase through autocomplete to confirm real readers are typing it.
  3. A paid data tool once you're running ads or scaling a series. Book #3 onward, when you're running paid AMS campaigns, or when one book is now a meaningful share of your monthly income — that's when Publisher Rocket's actual search-volume and competition data starts changing your decisions in ways autocomplete can't.

If you're publishing your first book and your total marketing budget is $200, skip the paid data tool. AI generator + autocomplete + the genre starter lists above will get you to a usable 7 keywords. Save the $199 for cover art or a copyedit.

Keywords vs categories: where they overlap, where they differ

Authors confuse these constantly. The relationship matters because Amazon explicitly tells you not to repeat category words in your keyword slots.

  • Categories are the shelves Amazon files your book on. You pick from a fixed taxonomy during KDP setup — up to 3 Amazon Store categories. Categories control which "Best Sellers in..." charts you can hit.
  • Keywords are the search phrases your book is indexed for beyond your category shelf. They control which reader queries surface your book on the search results page.

Where they overlap: Amazon uses both as ranking signals. A book in "Cozy Mystery" with keywords like cozy cat mystery small town is doubly anchored — categorically AND linguistically — to the cozy-mystery query cluster.

Where they differ: A category gives you shelf placement; a keyword gives you search-result placement. A book can rank #1 on its category bestseller list without ranking for any specific search query if its keywords are weak. And the inverse — strong keywords with the wrong categories — leaves money on the table because you're not appearing in category-browse traffic.

The category-keyword trick (extra category slots beyond the 3 defaults)

KDP gives you 3 Amazon Store category slots during setup. Most authors don't realize Amazon will assign your book to additional categories automatically when your keywords match category names that have an & or , qualifier — the so-called "keyword categories."

The qualifying list is published by Amazon and updates over time. Examples that have been on the list: cozies (qualifies for Cozy Mystery sub-categories), magical realism, dark fantasy, psychic, shifter (paranormal romance), lgbt (qualifies for LGBT fiction sub-categories), time travel, superheroes.

How to use it: when you're choosing your 7 keywords, check whether your subgenre is on Amazon's current keyword-category list. If shifter qualifies your paranormal romance for an additional Romance > Paranormal > Shapeshifter category placement, putting shapeshifter alpha romance in slot 4 earns you both search-query indexing AND a bonus category. One slot, two benefits.

Two cautions: (1) the list changes — what qualified in 2022 may not qualify in 2026. Check Amazon's current keyword help page or the Kindlepreneur category guide before relying on it. (2) the keyword must genuinely describe your book; gaming this with off-genre keywords is a content-guidelines violation.

For the full category strategy — the current qualifying-keyword list, how to request manual category additions via KDP support, and how to read category competition — see our Amazon Kindle categories guide. Pair the two strategies; don't run them separately.

What Amazon prohibits in keyword slots

Most guides cite "don't use Harry Potter" as the trademark example. That's anecdotal. Amazon's keyword help page lists the full prohibited categories — read these from the source:

  • Author names or contributors not associated with your book. This is the actual rule. Yes, that covers Brandon Sanderson, Colleen Hoover, and Stephen King — but also includes living authors with much smaller followings.
  • Brands or series titles you don't own. Harry Potter, Wheel of Time, Dune — but also non-book brands like Lego or Disney.
  • Subjective claims. bestseller, best, award-winning, top-rated, #1. Amazon doesn't let you self-anoint.
  • Time-sensitive statements. new, on sale now, coming soon.
  • Information duplicated in title, contributors, categories, or other metadata. Amazon's exact wording. This is the "don't waste slots on words you've already used" rule.
  • Generic terms. Words common to most books in a category (novel, book, fiction on its own).
  • Misspellings, common variants, or punctuation tricks. Amazon's search handles these automatically.
  • HTML tags and special characters. Will be stripped or rejected on save. Quotation marks are also listed by Amazon — they don't usually trigger rejection in practice, but they waste characters against your 50-char cap. Drop them.
  • Amazon program names. Kindle Unlimited, KDP Select, Prime — all prohibited.

Amazon calls this a "zero tolerance policy" and reserves the right to suppress your listing, remove the book, or — for repeat violations — take account-level action. Don't test it.

The post-launch keyword rescue workflow

Most KDP keyword guides assume you haven't published yet. The harder reality: your book is live, ranked 400,000+ BSR, and has been flat for 90 days. You don't need to republish — but before you start swapping keywords, confirm keywords are actually the problem.

Step 0 — Is it actually the keywords?

A flat-sales book has four likely culprits: cover, blurb, also-boughts, or keywords. Keywords are usually checked first because they're easiest to change, but they're rarely the largest lever. Diagnose in order:

  • Cover — if your AMS impressions are healthy (5,000+/week) but click-through is under 0.4%, the cover is failing. Readers see your book, they don't click. Fix the cover first; keywords won't help if nobody clicks.
  • Blurb — if click-through is fine but the product-page-to-sale conversion is under 8% (visible in AMS or estimated from impressions vs sales velocity), the blurb is failing. Readers click, then bounce. Rewrite the blurb before touching keywords.
  • Also-boughts — open your book page and scroll to the "Customers who bought this also bought" carousel. Are those books in your genre? If your cozy mystery's also-boughts are full of literary fiction or romance, Amazon has mis-classified your reader profile — usually because of mismatched categories or generic keywords, and categories are the lever to fix it.
  • Keywords — only if the cover, blurb, and also-boughts all check out, then keywords are the bottleneck.

The 6-step keyword rescue

  1. Set a 7-day baseline. Note your current BSR (best read at the same time each day), KENPC pages read if you're in KU, and any Amazon Ads impressions if you're running AMS. You can't measure a change without a "before."
  2. Check what queries your book is actually indexed for. Open an incognito Amazon tab. Search each of your 7 current keywords. If your book doesn't appear in the first 100 results for one of its own keywords, that slot isn't working — your phrase is too competitive or the indexing didn't take.
  3. Pull your AMS search-term report. If you're running any Amazon Ads — even a $5/day broad-match campaign — the search-term report shows every query that triggered your ad and which converted. Any converting query that isn't already in your 7 slots is a swap candidate. Filter the list before you slot anything: a slot needs to be 30–50 characters AND specific enough that you could realistically rank for it. Single-word converters like mystery or kindle book are too broad to use; 60+ character conversational queries are over the cap. The sweet spot is mid-length specific phrases that converted at least twice in 30 days. (If you're not running ads, a 30-day $5/day broad-match campaign on your existing 7 keywords pays for itself in the data you collect.)
  4. Identify your 2 weakest slots. The ones where you don't appear in the first 100 results, or where you appear on page 5+. Replace them with autocomplete-validated phrases or AMS-search-term winners. Keep the slots that are clearly working.
  5. Use Amazon autocomplete to validate replacements. Type the first 2–3 words of each candidate; the dropdown shows what readers actually type after that prefix. If autocomplete doesn't suggest your phrase or anything close, real readers aren't searching it.
  6. Wait 5–7 days, then re-check indexing for the new keywords. Repeat the cycle monthly if results are still flat.

What a real swap looks like

If your current 7 are all variations of one idea — cozy mystery / cozy mystery series / cozy mysteries for women / mystery novel / cozy crime / cozy mystery best / cozy mystery book — Amazon's semantic layer reads these as a single signal. You're using 7 slots to send 1.

A reallocation across the 3-role framework would look like the slate below. Treat these as candidates, not a finished list — run each through Amazon autocomplete (Kindle Store, incognito) before locking them into your dashboard.

  • Slot 1 (subgenre): cozy bakery mystery small town
  • Slot 2 (subgenre): amateur sleuth cat mystery recurring
  • Slot 3 (comp): culinary cozy mystery recipes included
  • Slot 4 (comp): knitting mystery cozy small town
  • Slot 5 (comp): holiday cozy mystery christmas village
  • Slot 6 (niche-modifier): clean cozy mystery no profanity
  • Slot 7 (niche-modifier): cozy mystery female protagonist over 50

Now you have 7 distinct discoverability signals — subgenre depth, comparable-author phrasing, and reader-niche modifiers — instead of one repeated keyword in seven outfits. The point isn't that any single phrase will rank on its own — it's that 7 distinct signals give Amazon's semantic layer 7 chances to match a reader query, versus 1 chance to match 7 ways.

Rules for the rescue cycle

  • Never change all 7 at once. Two or three changes per cycle is the max that lets you attribute impact. Change everything and you've learned nothing.
  • Log every change in a spreadsheet — date, slot number, old phrase, new phrase, BSR before, BSR after 7 days, BSR after 14 days, AMS impressions delta. Without a log, you're guessing in 60 days.
  • If 2–3 cycles (≈60 days) of keyword rescue don't move the needle, you're back at Step 0. The keywords probably aren't the limiting factor. Re-check cover, blurb, and also-boughts.

BSR context for 2026: per Kindlepreneur's BSR-to-sales calculator, Kindle Store BSR under 10,000 indicates strong daily sales; under 100,000 is steady; over 500,000 is infrequent. Indie-community analysts have flagged the older calculator coefficients as underestimating mid-range BSR by a meaningful margin, so 2024-era "BSR under 20K is great" rules of thumb are out of date — Amazon's catalog has grown faster than informal rules track. Amazon hasn't publicly disclosed changes to BSR math itself; the recalibration is to community estimation models.

A+ Content and non-English marketplaces

Two additional surfaces your book gets indexed on beyond the 7 keyword fields:

  • A+ Content — the rich-media section you can add below your book description in KDP. Text inside A+ Content modules is indexed (though weighted less than the keyword field itself). Don't keyword-stuff it, but writing the headings and module copy with your top reader-search phrases gives Amazon another signal. Books in series benefit most: a comparison module that says "If you liked [comp book / comp author phrasing], you'll like this series" earns indexing for those comp queries.
  • Non-English marketplaces (amazon.de, .es, .fr) — your 7 keywords apply globally. For English-language books selling into amazon.co.uk or .com.au, your existing keywords work because reader vocabulary is similar. For amazon.de or amazon.fr selling English-language books, autocomplete is meaningfully different — if a non-trivial share of your sales comes from these marketplaces, open them in incognito and run the alphabet-soup method on translated phrases your title or description already implies. The slots don't multiply per marketplace; you're choosing one set of 7 that performs across all of them.

Scaling keywords across a backlist (publisher view)

The 7-slot mechanics don't change at scale for a small press or a prolific indie with 20+ titles — but the workflow does:

  • Maintain a shared keyword spreadsheet per series and per imprint, so a new book's slots inherit proven phrases from sibling titles instead of starting from scratch each time.
  • Standardize the keyword-allocation template (the 3-role framework above) so every title in a backlist is filled out consistently — which makes A/B comparison across the catalog meaningful.
  • Pull the AMS search-term reports across the full catalog quarterly; converting queries on a backlist book are often the best keyword candidates for the next release in the same series.
  • For publishers shipping 20+ titles/year, the manual cost compounds fast. ManuscriptReport's white-label publisher services automate the keyword draft (alongside blurbs, categories, comp titles) so the editor's time goes to the half-dozen judgment calls per title, not the typing.

The full publisher AI-integration workflow — where keyword automation sits inside the broader editorial pipeline — is covered in AI integration in publishing workflows.

Frequently asked questions

How many keywords does Amazon allow on KDP?

Seven keyword fields per book, each capped at 50 characters. That's it for the backend. Your title, subtitle, contributor field, description, and category selections are also indexed, so the effective keyword surface is larger than 7 slots — but the 7 dashboard fields are the ones you directly control.

How long does it take for KDP keywords to update?

Typically 24–72 hours for re-indexing, occasionally longer for new books still in initial review. Don't judge impact for at least 5–7 days after a change — sales rank moves on weekly cycles, and same-day fluctuation is noise.

Can I change my KDP keywords after publishing?

Yes, anytime. Go to KDP Bookshelf → Edit Book Details → Keywords, save, and Amazon re-indexes. You don't need to upload a new manuscript or repeat any review steps. Keywords are metadata, not content.

Should I use single words or phrases in each slot?

Phrases, almost always. A reader searching just romance is browsing; a reader searching enemies to lovers workplace romance slow burn is shopping for something specific. Amazon's semantic layer (COSMO) understands the individual words inside a longer phrase, so a multi-word phrase gives you credit for both the exact match and the component concepts.

Should I repeat words from my title in my keyword slots?

No. Amazon's keyword policy explicitly tells authors not to duplicate information from the title, contributors, or categories. Amazon already indexes those — repeating them wastes a slot. Use the 7 fields for phrases you couldn't fit on the cover.

Are KDP keywords case-sensitive?

No. Amazon's search treats Cozy Mystery and cozy mystery as the same query. Capitalization is your choice for readability; it doesn't affect indexing.

What's the best free KDP keyword tool?

The combination of Amazon autocomplete (manual, free) and a focused AI generator (also free) for ideation. Our KDP Keyword Generator reads your synopsis and proposes 7 candidates in about 30 seconds — useful as a starting point you then validate against Amazon autocomplete. Neither replaces a paid data tool like Publisher Rocket when you need actual search-volume numbers, but both are sufficient for book #1.

How often should I update my KDP keywords?

There's no Amazon-mandated cadence. Most practitioners review keywords quarterly, or any time one of three events happens: sales soften meaningfully, a new comparable book becomes a bestseller in your niche, or you finish a paid-ads campaign and want to capture the data. After any change, wait 5–7 days before judging impact.


Related reading:


Keywords are one signal among many. If you're past the "fill the 7 slots and hope" stage and want a marketing system — keywords + categories + blurb + comp titles + ad copy + a 90-day launch plan — that's what ManuscriptReport's Book Marketing Report generates from your manuscript in about 10 minutes. $69 for the bundle that would otherwise take 20+ hours of research per book.

Turn your manuscript into a ready-to-market package

ManuscriptReport turns your full book into 20 launch-ready marketing assets — back-cover blurb, SEO keywords, comp titles, ad copy, and more.

Get Your Marketing Report →
FREE

Enjoyed this article? Subscribe for more + get a free marketing roadmap template.

Receive regular updates on marketing best-practices, AI shortcuts, and get our proven 4-phase marketing roadmap template for free.

Unsubscribe anytime.

Market your book like a bestseller

Get Your Report →