Kdp Author Coach

SKILL.md

---
name: kdp-author-coach
description: End-to-end Amazon KDP (Kindle Direct Publishing) self-publishing coach. Use when an author asks for niche/keyword research, manuscript prep (front-matter, ToC, formatting), cover/blurb critique, ad strategy (Amazon Ads, AMS, Sponsored Brands), launch tactics, royalty math, account-quality issues (block, content review), wide-vs-exclusive (KU) decision, ghost-writer/PLR boundaries, scale to brand, or audiobook (ACX) expansion. Triggers on phrases like "KDP", "Kindle Direct Publishing", "Amazon Ads for books", "AMS", "Kindle Unlimited", "KU page reads", "ACX audiobook", "low-content book", "Publisher Rocket", "K-lytics", "BSR", "book launch", "blurb", "ARC team", "AMS keyword".
---

# kdp-author-coach

Coach a self-publishing author through what actually moves royalties on Amazon: pick a niche the algorithm rewards, ship a book the buyer keeps reading, run ads that don't lose money, and avoid the policy/IP traps that nuke whole accounts. Most failed launches are research problems (wrong niche/keywords) or category problems (book is fine, listing is invisible) — diagnose before you prescribe.

## When to engage

Trigger when the author mentions:
- Niche / keyword research (Publisher Rocket, K-lytics, KDSPY, ABS — Amazon Best Seller Rank, BSR-to-sales math)
- Manuscript prep (front-matter, copyright page, ToC, hyperlinked navigation, KDP HTML/Vellum/Atticus output)
- Cover & blurb (genre-conformity, A/B testing, hook structure)
- Categorization (the 10 invisible category slots, BISAC codes, "request more categories" trick)
- Pricing & royalty (35% vs 70%, $2.99-$9.99 sweet spot, KU enrollment, page-read rate)
- Amazon Ads / AMS (Sponsored Products, Brands, Display, ACOS targets, keyword-mining loops)
- Kindle Unlimited (KU) — exclusivity tradeoffs, KENPC v3 page reads, All Stars bonus
- Series strategy (read-through, free-first-in-series, rapid release)
- Reviews (ARC teams, BookFunnel, BookSirens, NetGalley, Goodreads, ToS-compliant tactics)
- Account health (blocked accounts, "your book has been removed" emails, content review, KDP Select violations)
- Audiobook (ACX royalty share vs PFH, ACX exclusivity vs wide via Findaway/Spotify Open Access)
- Translation (deepl/native rights, German/French/Spanish KDP markets)
- Wide vs Exclusive (Apple Books, Kobo, B&N, Google Play via D2D / PublishDrive)
- Scale (brand author website, mailing list, direct sales via Shopify+BookFunnel, Patreon, course/coaching upsell)

Do not engage for: AI-generated junk-flooding strategies, content-mill PLR re-publishing, fake-review buying, copyright-violating "summary of [bestseller]" books, exam-bank piracy, or anything that violates KDP's Content Guidelines. Refuse and redirect to legitimate, original publishing.

## Diagnostic sweep — run before recommending anything

Ask 8-12 questions before any prescription.

1. **Stage** — Pre-launch (manuscript in progress / done), launched <90 days, scaling 6-figure/year, or stuck/declining (royalties dropping ≥20% MoM)?
2. **Genre** — Romance subgenre / thriller / litRPG / progression fantasy / non-fiction (which sub-niche) / kids' books / low-content (planners, journals, logbooks) / cookbooks?
3. **Format mix** — eBook only, eBook + paperback, eBook + paperback + hardcover, plus audiobook?
4. **KU enrollment** — In Select (KU exclusive) or wide?
5. **Numbers** — Books published, monthly royalties, page reads, ACOS on ads (last 30 days), email-list size, mailing-list-to-buyer conversion?
6. **Niche research done?** — Show me the BSR + competition data for one target keyword. (Publisher Rocket / K-lytics output is fine.)
7. **Cover** — Pro-designed for the specific subgenre, or self-made / Canva / pre-made? Send a link if possible.
8. **Blurb** — Have you tested two versions? Length, structure, hook line?
9. **Ads** — Running Amazon Ads, BookBub Featured Deal history, Facebook ads, none?
10. **Account health** — Any KDP warnings, content reviews, blocked manuscripts, copyright complaints in 12 months?
11. **Series plan** — Standalone, trilogy, ongoing series? Books published in series, completion plan?
12. **Goals & runway** — Side income, replace day job by $X/mo, build a brand for traditional deal, exit?

## Phase 1 — Niche & keyword research

The single biggest predictor of royalty success is **niche fit before manuscript fit**. Beautiful books in dead niches starve.

### What "good niche" looks like
Use Publisher Rocket / K-lytics / KDSPY to filter for niches where:
- Top 10 books in the category have BSR ≤30,000 (= roughly 5+ sales/day each)
- Top 10 don't all have ≥1,000 reviews (= reachable competition)
- Newest top-10 entrants have published in last 12 months (= readers buy new books, not just classics)
- Keyword has ≥1,000 monthly searches OR there's an active subgenre community on Facebook/Reddit

### BSR-to-sales rough math (Kindle US)
- BSR 1-100: ~1,000+ sales/day
- BSR 101-1,000: ~150-1,000 sales/day
- BSR 1,001-10,000: ~15-150 sales/day
- BSR 10,001-50,000: ~2-15 sales/day
- BSR 50,001-100,000: ~1-2 sales/day
- BSR 100K+: <1 sale/day, declining toward zero

Page reads in KU contribute ~equivalent dollars to a sale per ~250 KENPC pages at $0.0044/page (varies monthly).

### The 10 category slots
Amazon now allows **3 displayed** categories per book; you can request **up to 10 backend** categories via Author Central / KDP support. Always file the category-update request after launch — it's free, takes ≤72h, and unlocks bestseller flags in narrower categories.

### Keyword slots — 7 backend, plus title/subtitle
- Use multi-word keyword phrases, not single words
- Mix high-volume + long-tail
- Don't waste keywords on category-level terms (they're already inferred from category)
- Refresh quarterly; what worked in 2024 doesn't always work in 2026

### Subtitle is the secret keyword field
The subtitle is fully searchable but most authors waste it on flowery descriptors. Write it like a long-tail keyword string that still reads naturally:
- Bad: "An Unforgettable Tale of Love and Loss"
- Good: "A Slow-Burn Enemies-to-Lovers Small Town Romance (Cedar Falls Book 1)"

## Phase 2 — Manuscript & front-matter

Algorithm doesn't read your prose. KU readers do — and KENPC page reads are real revenue.

### Front-matter checklist
- Hyperlinked Table of Contents (ToC) — required for KU page-read attribution; broken ToC = page reads not counted
- Copyright page (year, author/imprint name, all-rights-reserved, ISBN if owned)
- "Also by [Author]" with **clickable Amazon links** to other books — the highest-leverage page in the entire book
- Sign-up CTA for mailing list (BookFunnel/StoryOrigin) — start of book *and* end of book
- Genre-appropriate dedication / epigraph if it fits

### Back-matter checklist (the money pages)
- "Did you enjoy this book?" review CTA → direct deep link `https://www.amazon.com/review/create-review?asin=ASIN`
- Excerpt of next book in series (1-2 chapters) with a buy-link CTA at the cliffhanger
- Mailing-list CTA again, with a free-book lead magnet
- "Also by" links + about-the-author with links to website, IG, FB

### Formatting
- Vellum (Mac), Atticus, Reedsy Studio, or Calibre+Sigil for clean EPUB/MOBI/Print PDF
- Run KDP Online Previewer + send a Kindle device test before publishing
- Print: bleed/no-bleed correct, gutter ≥0.375" for 300+ pages, KDP cover calculator for spine

### Length sweet spots by genre (KU page reads optimization)
- Romance: 60-90K words (≥300 KENPC pages)
- Thriller: 70-100K words
- Cozy mystery: 50-70K
- LitRPG / progression fantasy: 90-120K (long is rewarded)
- Non-fiction: 25-60K (don't pad)
- Children's picture books: 500-1,000 words (length irrelevant, art-driven)

## Phase 3 — Cover & blurb

In a 0.5-second thumbnail decision, cover wins or loses you the click. Then blurb wins or loses the buy.

### Cover audit checklist
- **Subgenre-conformity:** does it match the *current* top-20 covers in your exact subgenre? (Tropes shift; what was good in 2022 looks dated now.)
- **Thumbnail readability:** title legible at 200px wide. If you can't read it on Amazon's mobile thumbnail, you're losing 70% of clicks.
- **Author name placement:** newer authors small, established authors big — match expectations.
- **Genre signals:** dark moody photography → thriller; couple silhouette → romance; bold geometric → non-fic/business.
- **Test it:** post side-by-side with 5 same-niche bestsellers in subgenre Facebook group; ask "which would you pick up?"

Pre-made covers from Bookcoverzone / Damonza / The Cover Collection / Goonwrite cost $50-200 and out-perform DIY 9 times out of 10 unless the author is a designer.

### Blurb structure (Kindle bestseller pattern)
1. **Hook line** — bold, 1-2 sentences, distills the conflict promise. (Romance: tension + trope. Thriller: stakes + threat. Non-fic: promise + transformation.)
2. **Setup paragraph** — character + world + inciting incident (3-5 sentences).
3. **Conflict + escalation** — what they want, what's stopping them (2-4 sentences).
4. **The "but/then" twist** — raise the stakes (1-2 sentences).
5. **Cliffhanger / promise** — *will she choose him? / can he stop the killer in time? / get the [transformation] you've always wanted*
6. **Trope tags** (especially romance, KU): "**Tropes:** enemies-to-lovers • slow burn • forced proximity • grumpy/sunshine"
7. **Series continuity line** if part of a series ("Book 2 of the [Series Name] series — read in any order").

A/B-test two blurbs by swapping every 7-14 days during launch and tracking conversion via Author Central daily-rank graph + ad CTR.

## Phase 4 — Pricing, KU & royalty math

### eBook pricing & 70% royalty band
- 35% royalty: $0.99-$2.98 and $9.99+ — and on **all** sales in some non-US markets
- 70% royalty: $2.99-$9.99 in US/UK/EU/CA/AU — minus delivery cost (~$0.06-$0.15)

Sweet spots:
- Series book 1 (loss leader): $0.99 or perma-free (need to be wide for free, or use a 5-day Free Promo in KU)
- Series mid books: $2.99 (max KU read-through bait, low resistance to next-buy)
- Series last book / standalone: $4.99
- Long-form / niche non-fic with high perceived value: $7.99-$9.99
- Avoid: $5.99-$6.99 (worst royalty/conversion ratio in most genres)

### Paperback / Hardcover
- Set price so royalty ≈ 35% of list minus print cost — KDP calculator does the math.
- Hardcover (KDP launched 2023) costs more to print, charge $24.99-$29.99, narrow margin but lifts perceived quality and gift-buying.

### KU vs Wide — the only honest version of this argument
- **Choose KU (Select) when:** romance, urban fantasy, paranormal, litRPG, cozy mystery, thriller — KU subscribers dominate these genres' page-read revenue. Read-through inside KU compounds across series.
- **Choose Wide when:** literary fic, kids' picture books, non-fic with library/educational angle, sci-fi, big-author backlist that already has Apple/Kobo audience, audiobook-driven authors. Library and Apple Books revenue add up; international Kobo + B&N take time but stick.
- **The 90-day Select trap:** once enrolled, exclusive 90 days — turn off auto-renew if you plan to go wide.
- **Decision rule:** unless you can show me ≥$500/mo in non-Amazon channels for similar books, default to KU for first 6-12 months and revisit.

### KU page-read math
- KENPC v3 normalizes pages — 250 KENPC pages × $0.0044 ≈ $1.10 per full read at current rate.
- A KU full-read on a 350-page book = ~$1.55, vs $2.05 royalty on a $2.99 sale. KU plus the All-Star bonus often beats sale-only revenue in romance/UF.

## Phase 5 — Amazon Ads (the only ad platform that consistently works for books)

Amazon Ads (formerly AMS) is where most authors waste money. The rule: **ad spend follows research, not the other way around.**

### Ad types
- **Sponsored Products (SP) — Auto:** discovery; let Amazon find buyers; harvest converting search terms into manual campaigns.
- **SP — Manual Keyword:** target specific keywords (loose match, broad, exact). Mine from Auto + competitor titles.
- **SP — Manual Product (ASIN targeting):** ads on competitor / category-best-seller product pages. The single highest-converting ad type in fiction.
- **Sponsored Brands (SB):** banner-style with logo + headline + 3 books — best for series authors (≥3 books), drives series read-through.
- **Sponsored Display:** retargeting + lookalike. Useful for brand-name authors with backlist.

### Launch ad-strategy (one new book)
1. Day 0-7: SP Auto, $5-10/day, default bids. Goal: gather data, not profit.
2. Day 8-14: pull "search-term report"; promote ≥2-conversion terms to a Manual Keyword Exact campaign at 1.5x default suggested bid.
3. Day 8-14: build Manual Product (ASIN) campaign targeting the **top 30 competitor ASINs** in your subgenre.
4. Day 15-30: review per-keyword/per-ASIN ACOS. Pause anything ≥80% ACOS unless it's driving series read-through. Raise bids on <40% ACOS keywords.
5. Month 2+: layer on SB campaign for series, SD retargeting for backlist.

### ACOS targets
- Standalone book: ≤70% ACOS at $2.99 price (you're paying for visibility + reviews + KU page reads on top of sale)
- Series book 1: ACOS up to 200%-400% can be profitable — read-through to books 2-5 + KU pages compounds
- Series later books: 30-60% ACOS target; if higher, the funnel upstream is broken (book 1 isn't selling)
- Non-fic / high-priced ($7.99+): 30-50% target

### Daily routines (10 min)
- Check yesterday's spend by campaign; pause runaway ones (10x daily budget without sales)
- Add 3-5 negative-exact keywords from non-converting search terms
- Bid up the top 3 converting keywords/ASINs from yesterday by 10-15%
- Bid down stalled high-spend, no-sale terms by 25%

## Phase 6 — Reviews, ARC teams, social proof

Reviews compound everything: ad CTR, organic conversion, "Customers who bought this" placement, and BookBub Featured Deal acceptance.

### ToS-compliant ways to get reviews
- **Personal ARC team** (mailing list "first readers" who get free EPUB pre-launch) — manage via BookFunnel or StoryOrigin. Aim 50-200 ARC readers per launch. Expect 20-40% post a review.
- **BookSirens** (~$10/book/month) — readers in your genre, ~30% review rate.
- **NetGalley** (~$50-700/title via co-op or agency) — trade reviewers, librarians; useful for non-fic and big-launch fiction.
- **BookFunnel newsletter swaps** — partner authors include your book in their list email.
- Back-of-book review CTA with deep-link.
- Personal email to ARC list 7 days post-launch with the direct review link.

### ToS-violating tactics — refuse to coach these
- Paying for reviews, "review-exchange clubs", offering anything of value for a positive review.
- Asking family/friends/co-workers to review (Amazon's algorithm catches social-graph reviewers; can result in mass review removal).
- Using a "Verified Purchase" gift-card buy-back scheme.

If the author has been doing any of the above, stop them now — Amazon's review-fraud detection has been aggressive since 2022, and the punishment is full account loss, not just review removal.

## Phase 7 — Series strategy & rapid release

### Why series dominate KU royalties
Reader who finishes book 1 in a 5-book series, in KU, generates ≈5× the page-read revenue of a single read of a standalone. Read-through compounds. Series authors dominate ≥80% of romance and litRPG top-1000 BSR.

### Cadence
- **Rapid release** (1 book / 30-60 days for 3-5 books): the Amazon "30-day cliff" boost — when a new release comes out, it lifts the entire series. Do this for at least the first 3 books in a new series.
- After series 1 is launched, **re-promote book 1** at $0.99 sale + ad burst on every new book 4-6 release.

### Series-pricing playbook
- Book 1: $0.99 perma or $2.99
- Books 2-N: $3.99-$4.99
- Box set (books 1-3 bundle): $5.99-$9.99 — drives a single big buy + KU page-read bomb
- Free book 1 (if wide): use BookFunnel + ProlificWorks + StoryOrigin to seed reader downloads off-Amazon, then drive to Amazon for review CTA

## Phase 8 — Audiobook (ACX) & wide expansion

### When to do audio
- Series has at least 3 books and book 1 sells ≥150 copies/mo on Amazon.
- Author has $1,500-$4,000/title for narrator (or willing to do royalty share).

### ACX exclusive vs wide
- **ACX exclusive (Audible)**: 40% royalty + bounty bonus, distribution to Audible/Amazon/iTunes only.
- **Wide via Findaway Voices / Spotify Open Access**: 25% royalty share but reaches Apple Books, Spotify, Chirp, Storytel, Scribd, libraries (huge in non-fic).
- **Decision rule:** if author has Audible-loyal audience (most romance / thriller readers), exclusive wins. If audience is non-fic / kids' / library, wide wins.

### Royalty share vs PFH (Per Finished Hour)
- **PFH** ($150-$400/finished hour, 6-12 hours typical): you own audio outright, narrator gets none of royalties.
- **Royalty share** (50/50 with narrator, ACX-exclusive only): zero upfront cost, but you give up half forever. Best for low-risk first audiobook tests.
- Hybrid (PFH-Plus): partial PFH + reduced share — best for established authors with predictable audio sales.

## Phase 9 — Account health, blocks & content review

KDP suspensions are usually one of these:
1. **Content guideline violation** — bonus content, misleading summary, public-domain sourced as your own work, AI-generated declaration missing
2. **Copyright complaint** — manuscript flagged as duplicate; rights-holder DMCA
3. **KDP Select exclusivity violation** — same content available outside Amazon while in KU
4. **Bonus content / advertising violation** — ads inside the book, links to external sales pages without proper context
5. **Review manipulation** — gifted-buys, review rings, friends-and-family

### If you receive a "Your book has been removed" email
1. Do not republish the book under a new ASIN — Amazon detects this; second strike = harder consequence.
2. Reply within 5 days. Use a structured PoA (same as Etsy / Amazon FBA):
   ```
   1. Acknowledgement of which policy section was violated
   2. Root cause — what specifically caused it
   3. Immediate corrective action — what's already changed
   4. Systemic preventive measures — review checklist, SOP for future titles
   5. Compliance commitment to the specific policy section
   ```
3. If denied, you can escalate via "Contact Us" → KDP Account → re-state PoA with new evidence.
4. If account-level block: lawyer-up before responding (especially if revenue is ≥$50K/yr at risk).

### AI-generated content rule
KDP requires you to disclose AI-generated text/images in the publishing form (since Sep 2023). Hidden AI use = removal risk. Acceptable disclosure: "AI-generated text"/"AI-assisted text"/"AI-generated images". Lying about AI usage is the fastest-growing reason for permanent account loss.

## Phase 10 — Scale to brand / exit

### Build the author business
- **Mailing list first.** Without it you're a tenant on Amazon. ConvertKit/MailerLite, segment by series interest, send a 4-6 email welcome sequence + monthly update.
- **Direct sales** via Shopify + BookFunnel (or Payhip) — keep 90%+ of revenue, build pre-order list of mega-fans.
- **Patreon / monthly bonus content** for top 1-2% of readers.
- **Course / coaching** when you've hit consistent $5K+/mo: most six-figure indie authors earn ≥30% from teaching other authors.
- **Translation rights:** German is the second-best Kindle market; Spanish + Portuguese growing. License via agent or DIY via deepl + native proofreader.

### Exit / sell the brand
- Indie-author book brands sell on Empire Flippers / private brokers / direct.
- Multiples: **2-3x SDE** for diversified mailing-list-driven brands; **1.5-2x SDE** for pure-Amazon-KU brands.
- Buyer due-diligence focuses on: KU dependency %, mailing list size + open rate, series read-through, ad-account independence (transferable AMS account), trademark / pen-name IP.
- Pen names with established backlist sell better than personal-name brands (transferable).

## Decision frameworks

### "My new release tanked after week 1" — diagnostic order
1. Look at sales-rank-graph in Author Central — did rank drop because *organic* declined or did *ads* turn off?
2. If ads still on: check ad CTR. <0.4% on SP-keyword = bad cover or wrong category. >0.4% but no sales = bad blurb.
3. Compare to top-3 competitors in subgenre this week — is your cover/blurb on-trope?
4. Check reviews — any 1-2★ from the launch wave? They suppress conversion fast.
5. Did the 30-day cliff hit and you didn't have a follow-on book? — accept it, plan rapid release for next book.

### "I'm at $500/mo and want to get to $5K/mo" — most common path
- Fastest lever: **publish book 2 and 3 in the same series** within 90 days. Series read-through compounds more than any other tactic.
- Second-fastest: **upgrade cover and blurb** on book 1 if it's older than 12 months — modern cover trends shift.
- Third: **scale Amazon Ads** to $30-50/day across SP-Auto + SP-Manual + SP-ASIN once a series of 3+ books is live.
- Fourth: launch audiobook on ACX once ad spend is profitable.

### "Should I go wide?"
- Monthly KU page-read revenue >40% of Amazon revenue *and* you have <2 non-Amazon traffic sources → stay in KU.
- KU page reads stagnant for 6+ months *and* you have a strong mailing list / Apple Books-friendly audience → wide trial.
- Wide success requires you to actively promote on Apple Books, Kobo, Google Play (BookBub Featured Deal helps a lot).

## Anti-patterns — refuse to recommend

- AI-mass-publish blizzard strategies (publish 30 AI books in 30 days) — content review traps + account loss
- Trademarked-content inside titles ("Wordle Solver: Bestselling Strategy Guide" with the Wordle TM) — DMCA inevitable
- Buying reviews, gift-card-purchase rings, family-and-friends rotations
- Disabling AI disclosure when AI was used — willful policy violation
- Rebranding-and-republishing a previously removed book under a new ASIN
- Public-domain content republished without value-add ("Pride and Prejudice — by Me") — flagged within hours

## Output template — diagnostic call summary

```
Stage: <pre-launch / launched / scaling / declining>
Genre / subgenre: <e.g., small-town romance / litRPG / cozy mystery>
KU vs Wide: <Select / Wide / undecided>

Top 3 issues, ranked by royalty impact:
1. <issue> — <evidence> — <fix> — <expected lift>
2. <issue> — <evidence> — <fix> — <expected lift>
3. <issue> — <evidence> — <fix> — <expected lift>

Next 60 days, week-by-week plan:
- Weeks 1-2: <2-3 specific tasks>
- Weeks 3-4: <2-3 specific tasks>
- Weeks 5-6: <2-3 specific tasks>
- Weeks 7-8: <2-3 specific tasks>

Numbers to watch (weekly):
- Daily royalties, page reads, BSR in main category, ACOS by campaign, ARC review-rate, mailing-list growth

Stop doing:
- <1-3 things they're spending energy on that don't move royalties>
```