Royalty Optimization: The Hidden Levers in KDP Reports

Most authors treat KDP royalties as a fixed function of price and units. They are not. There are at least seven levers in the KDP reports that move royalty per book without changing a word of the manuscript — and three of them are buried deep enough that most authors never see them. Here is where to find them, what to look for, and how to act.

Lever 1 · The 70/35 royalty cliff and where it costs you

The headline KDP rule: ebooks priced $2.99-$9.99 get 70 percent royalty (minus delivery fees); ebooks priced below $2.99 or above $9.99 get 35 percent. Most authors know this. Most accounts still leak money on it.

The leak: list price changes for promotions never roll back automatically. A free run that drops a book to $0.99 for 5 days will end with the price restored to $4.99 by KDP — but only if you set it up via Kindle Countdown Deals. If you used manual price edits, the price stays at $0.99 indefinitely. We see this in roughly 1 in 8 accounts on first audit.

Where to find it: open the Pricing & Royalties report(KDP Reports → Royalties → Pricing & Royalties export, CSV). Sort by list_price_marketplaceascending. Anything under $2.99 that should not be there is a 65 percent revenue cut sitting in plain sight. We see authors recover $200-$800/month from this single check.

The opposite cliff is rarer but real: ebooks pushed above $9.99 for marketing reasons (typically box sets) drop back to 35 percent. Box sets above $9.99 are reasonable; individual novels at $11.99 almost always pay less royalty than the same novel at $7.99 once volume effects shake out.

Lever 2 · Delivery fees on KU-eligible ebooks

The 70 percent royalty plan deducts a per-megabyte delivery fee — $0.15/MB on amazon.com, €0.12 on most EU marketplaces. For a 280-page novel exported from Vellum, file size is typically 2-4 MB and the fee is a rounding error.

For non-fiction with images, illustrated children’s books, and cookbooks, file size routinely exceeds 30 MB and delivery fees can take 10-20 percent of royalty. The lever:AVIF or modern WebP compression on cover and interior images can cut file size by 30-60 percent with no visible quality loss on Kindle hardware.

Where to find your file size: KDP Bookshelf → click the book → Kindle eBook Content tab. The “File Size” line shows the rendered KFX file. If it is over 5 MB on a fiction title or over 25 MB on illustrated, run a re-compression pass and re-upload. The royalty bump is automatic on the next sale.

Lever 3 · KENP page count vs reading time

KDP Select payouts are based on KENP — Kindle Edition Normalized Pages — not your manuscript’s actual page count. KENP is Amazon’s standardized page metric: roughly one KENP per 187 words at default Kindle font and spacing. The lever most authors miss: KENP rewards readability, not length directly.

Specifically: chapter breaks, dropped capitals, scene-break ornaments, and pull-quotes all add KENP. A novel with 78,000 words and aggressive whitespace formatting registers around 420 KENP. The same 78,000 words exported with default Vellum settings registers 380 KENP. That is a 10 percent KU royalty lift for a styling change you make once.

Where to verify: KDP Reports → KENP Read Pages. Compare lifetime KENP-per-unit-borrowed across your books. A 2024-published book showing 280 KENP/borrow vs a 2023-published same-length book showing 410 KENP/borrow is a formatting regression, not reader behavior.

We have seen authors lift KU income 8-12 percent by re-uploading existing books with corrected formatting. KDP allows unlimited interior re-uploads on a published book; the new file propagates within 24-48 hours.

Lever 4 · Marketplace-specific list prices

KDP defaults non-USD list prices to a USD-converted equivalent plus marketplace VAT. This is almost always the wrong price.

On amazon.de, the round-number sweet spot for ebook fiction in 2026 is €3.99 or €4.99. A USD-converted price like €5.49 sits above the price point readers are used to seeing, even though the royalty per copy is higher. Test it on one title: the lift on units can more than offset the lower per-unit royalty.

The round-number price points readers in each store are used to, and worth testing against a converted price:

• amazon.com— $3.99 is the floor for 70 percent royalty fiction; $4.99 is the median sweet spot; $6.99 is the highest price that still hits the impulse-buy bucket on most sub-genres.
• amazon.co.uk— £1.99, £2.99, £3.99 are the conversion plateaus. £4.99 drops significantly for fiction.
• amazon.de— €3.99 and €4.99 are plateaus. €5.99 cuts conversion by ~25 percent versus €4.99.
• amazon.com.au — A$4.99-A$6.99 is the comfortable range. Below A$4.99 looks suspicious to readers who expect to pay more for fiction.

Where to set: KDP Bookshelf → book → Kindle eBook Pricing → Individual marketplace prices. Manual entry per marketplace, then confirm royalty band per row. Plan 30 minutes per book. The lift compounds — list price changes propagate within 2 hours and start affecting all subsequent sales.

Lever 5 · Print royalty math on color interiors

KDP Print’s 2024 expansion of color print options introduced a tier most authors haven’t reconsidered: “Standard color” on premium paper. Production cost is significantly lower than “Premium color” — for a 100-page interior, the gap is roughly $1.20 versus $4.40 — and quality is acceptable for most non-photography books.

Where to check: KDP Bookshelf → Print Title → Print Cost. If your color interior shows a print cost above $5 on a sub-200-page book, you are likely on premium color when standard would do. Switching the option is one dropdown in the Paperback Content tab.

Royalty lift varies by retail price but is typically $2-$4/copy. For a low-volume but steady-selling children’s book, this is the difference between the format paying for itself and not.

Lever 6 · Hardcover format halo

Books available in hardcover sell more paperbacks than books without a hardcover edition. The halo is about 7-12 percent on paperback units when a hardcover is published — even when the hardcover sells almost no copies itself. Amazon’s product page treats hardcover as a completeness signal.

KDP’s 2024 hardcover format expanded to most marketplaces in 2025. If you have a paperback selling steadily and have not added a hardcover, the lever is straightforward: hardcover production cost is $2.00 + $0.012/page; price the hardcover at roughly 2× your paperback list. Hardcover royalty is thin but existent; the paperback halo is the actual return.

Where to check whether you have a hardcover edition: KDP Bookshelf → click book → format tabs at top. If only Kindle and Paperback are present, hardcover is a 30-minute setup.

Lever 7 · KDP Select exclusivity audit

KDP Select gives you KU royalties and the price-promotion tools — but locks the ebook to Amazon for 90-day terms. The exclusivity cost is invisible to most authors because the alternative revenue (Apple Books, Kobo, Google Play) is hypothetical.

For most fiction authors with no existing audience on other platforms, the math favors KU. For authors with a mailing list or a strong .epub audience already, the calculation flips.

The diagnostic: in KDP Reports → KENP Read Pages → compare against Sales. If your KU royalty per book per month is below your sales royalty by 50 percent or more, KU is not paying for the exclusivity it costs you. Especially for non-fiction and series fiction with strong read-through, this gap can be large.

Action: KDP Select renews automatically on 90-day cycles unless you opt out. Set a calendar reminder for 14 days before each renewal date — you only need that window to decide and act.

The unsexy summary

Royalty optimization is not flashy. It is reading three reports a month and acting on what they say. The seven levers above represent something like 20-40 percent of monthly royalty for a typical author who has never specifically looked at any of them. The compounding effect is significant — and unlike ad spend, every gain here is permanent.

Most of these are spreadsheet-level work. SelfPub Go automates the audits — surfaces the leaks, computes the lift, and flags renewals — but a determined author with two hours and the patience to read CSVs can do all of it manually. The results are the same. The choice is yours.