Authors & Publishing
How to Write and Publish an eBook in 2026: KDP, Marketing, and a 2027 Launch Plan
Pre-order, royalty, A+ Content and review-policy rules were re-checked against Amazon KDP Help on 26 October 2026; the other sources were checked on 25 October 2026. Platform rules change; each rule below links to its source.
Short answer. To write and publish an eBook: pick one reader and one promise, check that readers already buy similar books, outline and draft it, get an outside edit, package it (cover, file, title, description, seven keywords, up to three categories), choose KDP Select or wide distribution, price it knowing what a sale actually pays, and launch to people you can reach. KDP does not require an ISBN for an eBook [S6], accepts KPF, EPUB and DOCX [S3], and lets you schedule a pre-order up to 18 months ahead if you create it on or after 2 September 2026 [S1]. The part most guides skip is the order of the decisions, and when to wait. That is what this guide is built around.
How to use it. Start with the decision map below. Use the three tools (royalty calculator, launch-date planner, readiness scorecard) when you reach them. Every rule is labeled so you can tell what Amazon requires from what is only advice:
- FACT A platform rule or number, with a source link and a checked date.
- RULE OF THUMB Common practice. Useful, not guaranteed.
- EXAMPLE A worked, clearly hypothetical illustration. Not a customer result.
- YOUR DECISION A choice that depends on your goals. We say what to weigh, not what to pick.
If you are mid-project, jump to your stage: I have an idea · I’m drafting · My manuscript is finished · I’m querying agents · I’m ready to self-publish · I just published · I’m planning Q4 2026 or 2027.
eBook journey map
Nine stages from idea to a 2027 plan. Each stage names the decision to make there and links to that part of this guide.
- IdeaIs an eBook the right format?
- ValidationDoes a reader already buy books like this?
- ManuscriptCan you keep a daily draft habit?
- ReadinessDoes a stranger finish it and say what it is about?
- PackagingCover, file, title, seven keywords, categories
- KDP and distributionPrice, royalty option, Select or wide
- DiscoveryStore page, ads, Author Central, reviews rules
- First readersWho are your first 100, and where do they come from?
- Post-launch and 2027What do you measure, and when do you release?
Is an eBook the right format for you?
YOUR DECISION An eBook is a good fit when the book is complete at a length readers will finish, you want to control price, timing and covers, and you can reach readers yourself. It is a weaker fit when your main goal is a traditional publishing deal, a printed book on shelves, or a very large illustrated layout.
Three questions settle most cases:
- What do you want the book to do? Earn royalties, build credibility for a business, serve a list you already have, or reach a traditional publisher. Each points to different choices later.
- Who will read it? If you can name them, you can find them. If you cannot, stop here and fix that first (next section).
- Can you support it after launch? A published eBook with no plan to reach readers is the common failure, not bad writing.
Not sure whether to query agents, self-publish, or revise first? Use the decision tree. There is no universal right answer.
Choose one reader and write a one-sentence promise
RULE OF THUMB Write this sentence before the outline:
For [specific reader], this book helps them [outcome or experience] without [the usual pain].
Non-fiction promises an outcome (“a first draft in 60 days without an outline you abandon”). Fiction promises an experience (“a slow-burn mystery in a small town where nobody is who they claim”). If three people outside your circle cannot repeat it back to you, it is not specific enough.
Validate the idea in an afternoon
RULE OF THUMB You are looking for evidence that readers already buy books like yours, not for a precise market size.
- Search the Kindle Store with the words your reader would type. List the top ten results: price, length, cover style, publication date.
- Read the one-, two- and three-star reviews. They tell you what readers wanted and did not get. That gap is your angle.
- Write down what your book does that those do not.
- If nothing like it exists, ask honestly whether that is an opening or a sign of no demand.
Reader promise and validation worksheet
Write the one-sentence promise, then record the top ten similar books you found in the Kindle Store. Nothing you type leaves your browser.
Top ten similar books
| # | Title | Price ($) | Pages | Year |
|---|---|---|---|---|
| 1 | ||||
| 2 | ||||
| 3 | ||||
| 4 | ||||
| 5 | ||||
| 6 | ||||
| 7 | ||||
| 8 | ||||
| 9 | ||||
| 10 |
Outline and draft with a system you can keep
RULE OF THUMB Outline in the shape of your reader’s journey, not your own knowledge. For non-fiction, one problem per chapter. For fiction, a one-page structure and a scene list.
EXAMPLE Hypothetical arithmetic, not a result: a 40,000-word guide at 500 words a day is about 80 writing days; at 1,000 words a day, about 40. Pick a daily number you can actually hit, then stop editing while you draft.
Draft first, edit later. Mixing them is the most common reason first drafts stall.
Editing, and the moment a manuscript is actually ready
FACT KDP does not require an edit. That is a quality decision, not a platform rule.
There are three different jobs, and they happen in this order:
| Level | What it fixes | When |
|---|---|---|
| Developmental edit | Structure, pacing, argument, character arcs | First |
| Copy edit | Sentences, consistency, style | After structure is settled |
| Proofread | Typos and formatting slips | Last, on the final file |
RULE OF THUMB Many first-time authors pay for a proofread and skip structure, which fixes the cheapest problems first. Get outside readers who are not close to you to look at structure before you pay anyone.
Your manuscript is ready when: a reader outside your circle can finish it and tell you what it is about, the structure no longer changes when you read it, and the remaining edits are line-level only.
FACT KDP asks you to tell it if text, images or translations are AI-generated, even if you edited them afterward. Content you created yourself, where tools only helped you edit, refine, error-check or brainstorm, is “AI-assisted” and does not need to be disclosed. You are responsible for rights and policy compliance either way. [S7]
Query, self-publish, or revise? A decision map
Query, self-publish or revise: decision tree
Answer up to three questions. It points to a next step; it does not predict an outcome, and there is no universal right answer.
The table above gives the same logic in text.
YOUR DECISION There is no universal answer. These are the questions that change it:
| If this is true… | It points toward |
|---|---|
| You want a traditional publisher and an agent | Query seriously first. Many agents expect unpublished work, so check each agent’s guidelines before you publish anything yourself. |
| You want speed, control and your own pricing | Self-publishing, with a real edit and cover. |
| Your manuscript still changes when you read it | Revise before any path. |
| You are querying and getting no requests at all | Look at the query package (letter, synopsis, comparable titles, opening pages) before the book. |
| You are getting requests but then passes | The package may be fine; the issue could be fit, timing or the manuscript. Do not assume which. |
| You are undecided | Set a decision date in advance, query to that date, and keep the self-publishing plan ready. |
If you are querying and unsure about your package. A second pair of eyes on consistency across your letter, synopsis, comparable titles and sample pages often finds drift after many rewrites. BMT’s Query Package Consistency Check ($99) does that review. It is fit-gated: we confirm it suits your stage before you pay, and it does not promise representation. Ask about it here.
If you are unsure the manuscript itself is ready. That is a different question from the query package. BMT offers a Manuscript Readiness conversation to work out what stage your draft is at before you commit to a path. There is no public price; start with a short fit check.
Packaging: cover, file, title, description, keywords, categories
Cover
RULE OF THUMB Readers judge a cover at thumbnail size. Check yours at the size of a search result: can you read the title, is the genre obvious, does it sit comfortably next to the top ten you listed earlier? Hire a designer who works in your genre if you can. A clean simple cover beats an ambitious one that fails at small size.
Cover thumbnail test
Choose your cover image to see it at three small sizes. The file is read in your browser and is never uploaded.
No image chosen yet.
At each size ask: can you read the title, is the genre obvious, does it sit comfortably next to your top ten?
File format
FACT KDP’s help page lists Kindle Package Format (KPF, made with Kindle Create), EPUB and Word (.doc/.docx) among accepted eBook manuscript formats; it presents KPF as the most reliable way to avoid formatting trouble, asks you to validate EPUB files in Kindle Previewer, and warns that complex Word formatting may not convert well. MOBI is no longer accepted for fixed-layout eBooks as of March 2025. [S3]
RULE OF THUMB Mostly text: a clean DOCX or EPUB is usually enough. Images, tables or special layout: use Kindle Create or a formatter. Either way, preview in Kindle Previewer before you upload.
| Format | Good for | Watch for |
|---|---|---|
| KPF (Kindle Create) | Layout-heavy books, fewest surprises on Kindle | Kindle-specific; not a general EPUB for other stores |
| EPUB | Reflowable text; also usable on other stores | Must meet Kindle guidelines; check in Previewer |
| DOCX | Simple text-heavy books | Complex formatting may convert badly |
Title, subtitle, description
FACT The title field must hold only the book’s actual title, as it appears on the cover. The subtitle is optional, and title plus subtitle must be under 200 characters. Descriptions cannot contain contact details, URLs, reviews or testimonials, requests for reviews, price or time-sensitive information. [S6]
RULE OF THUMB The first two lines of the description do the work. Say the genre, the situation and the stakes (fiction), or the reader and the promise (non-fiction). Avoid vague openers like “This book will change your life.”
Keywords and categories
FACT You can enter up to seven keywords or short phrases. KDP advises against repeating words already in your title or categories, against subjective claims, time-sensitive words like “new” and program names like “KDP Select.” [S5] You can choose up to three categories at setup. [S6]
RULE OF THUMB Choose categories by where books like yours actually sit, and keywords by what your reader types.
Metadata worksheet
Draft title, subtitle, description, seven keywords and three categories. The checks reflect KDP Help pages S5 and S6; KDP does the final validation.
Price and royalties: what a sale actually pays
FACT KDP offers a 35% and a 70% royalty option. On Amazon.com the 70% option applies to list prices from $2.99 to $12.99 (the $12.99 ceiling took effect 7 July 2026, up from $9.99). The 35% option runs from $0.99 up to $200, with the minimum depending on file size: $0.99 under 3 MB, $1.99 for 3 MB to under 10 MB, $2.99 for 10 MB or more. [S2]
FACT The 70% formula is 70% x (list price – applicable VAT – delivery costs). Delivery costs on Amazon.com are $0.15 per MB of file size, rounded up to the nearest kilobyte. The 35% formula is 35% x (list price – applicable VAT) with no delivery charge. Outside the 70% territories, sales earn 35%; for Brazil, Japan, Mexico and India the book must also be in KDP Select to earn 70%. [S4]
EXAMPLE Hypothetical, US sale, no VAT: at $4.99 and a 1 MB file, the 70% option pays 0.70 x ($4.99 – $0.15) = $3.39; the 35% option pays $1.75. At $4.99 and a 5 MB image-heavy file, delivery is $0.75 and 70% pays $2.97. At $2.99 and 1 MB, 70% pays $1.99.
KDP royalty calculator (Amazon.com, US dollars)
Compare the 35% and 70% options at a price and file size. Estimates, not promises.
Assumes a US sale with 0% VAT. 70% requires a list price of $2.99 to $12.99 and subtracts $0.15 per MB (file size rounded up to the nearest KB). 35% has no delivery charge but a size-based minimum price. Sales in some territories earn 35% regardless; for Brazil, Japan, Mexico and India the book must also be in KDP Select for 70%. Kindle+ page-read income depends on a monthly fund and is not modelled. Sources: S2, S4.
YOUR DECISION Price to what your reader expects in your genre (check your top-ten list), not to a number that feels generous. A very low price rarely makes up for a weak page.
KDP Select or wide distribution?
FACT KDP Select enrollment lasts 90 days and renews automatically unless you turn renewal off. During enrollment the Kindle eBook can be distributed only through KDP and public libraries; print and other formats can be sold elsewhere. You can cancel within three days of the enrollment period’s start, except for a pre-order, which cannot be cancelled until it is released. [S8]
FACT Select titles are in Kindle+ and are paid from the KDP Select Global Fund based on Kindle Edition Normalized Pages (KENP) read. The per-page rate changes every month, so no fixed rate can be quoted. [S10] You can also run a Free Book Promotion for up to five days in each 90-day period, with no royalties while the book is free. [S9]
| Option | KDP Select | Wide |
|---|---|---|
| Where the eBook sells | Kindle Store (and public libraries) | Kindle plus other stores |
| Extra income | Kindle+ page reads, Free Promotion days | Other stores’ sales |
| Control | Exclusive for 90-day terms | Per-store setup |
| Main risk | Locked to one store | More setup and reporting |
YOUR DECISION If your readers borrow heavily, Select may fit. If other storefronts matter to your audience, stay wide. With one book, either choice can be reversed after a term. We have not verified third-party aggregator fees in this guide; check each store’s current terms.
Pre-orders and release dates
FACT For pre-orders created on or after 2 September 2026, you can schedule the release up to 18 months ahead (up to one year for pre-orders created before that date). Release time is in GMT. You must upload your final manuscript and submit the pre-order at least 72 hours before release; book details and manuscript cannot be changed inside that 72-hour window. [S1] [S11]
FACT Missing the deadline cancels the pre-order and counts against your account. KDP’s page says the first cancellation or missed deadline is a one-time courtesy; the second removes pre-order privileges for six months across your whole catalog. Pre-orders created on or after 2 September 2026 allow up to two postponements within six months of the original release date. [S12]
RULE OF THUMB Treat 72 hours as a hard floor, not a plan. Submit at least one to two weeks early. Use a pre-order when the cover and description are final and you have people to point at the page. Do not use one if the manuscript is not finished.
Author Central, A+ Content, and your store page
FACT Author Central lets you add books to an author page, add a biography, check your Amazon Best Sellers Rank and customer reviews, add editorial reviews and see BookScan sales information (the last two are US only). You need a KDP account and a published book to claim the page. [S13]
FACT A+ Content is available for published KDP books and for pre-orders; compliant content appears within eight business days and shows in the “From the Publisher” section, which most readers must scroll to see. Customer reviews cannot be used in A+ Content, and quotes from individuals or customers are not allowed; up to four quotes from well-known publications or public figures, with the source cited, are permitted. [S14]
RULE OF THUMB A+ is for what the description cannot show: a short series map, a clean author note, a comparison table. Do not expect it to rescue a weak cover or description.
Reviews and advance readers: what is allowed
FACT KDP says you may provide free or discounted copies of your books to readers, as long as you do not require a review in exchange or attempt to influence the review. Offering anything other than a free or discounted copy, including gift cards, will invalidate a review. Amazon’s Community Guidelines also say readers can review free advance copies from authors, but authors cannot require a review in exchange or try to influence the reader’s opinion. [S15]
FACT Amazon’s Community Guidelines tell reviewers not to post reviews on products from sellers, authors, brands or artists they have a personal connection with, and their conflict-of-interest rule covers products offered by friends, relatives, employers and business associates. Amazon describes a message such as “Please share your honest feedback experience with our product by leaving a review!” as an example of an acceptable seller request. [S15]
RULE OF THUMB If you use advance reader copies, eligible readers may choose to leave an optional, honest review. Do not require, incentivize, pressure, or influence a review or its sentiment. Do not solicit Amazon customer reviews from friends, family, or other close personal connections. Expect that many readers will not leave a review. Never pay, never script.
Amazon ads: the basics and the arithmetic
FACT KDP offers Sponsored Products (pay-per-click ads on search results and product pages, recommended as the starting point) and Sponsored Brands (which require a pen name with three unique titles in Author Central). You set your own bid and budget; KDP’s page states no minimum. Every ad goes through moderation, usually within 24 hours and up to three business days. [S16]
EXAMPLE Hypothetical arithmetic with stated assumptions, not a benchmark: at a $0.50 click and 10% of clicks buying, each sale costs $5.00 in ads. If a sale pays you $3.39, that is a $1.61 loss per sale unless it leads to more reading or a next book. Run your own numbers before spending.
RULE OF THUMB Ads usually make sense after you have a tested store page and, ideally, a series. Start small, cap daily spend, and stop what does not convert.
Author website, email list, social, video and podcasts
Author platform stack
What Amazon owns, what you own, and what carries reach. Build the bottom layer first.
Same diagram as text
- Amazon owns: your store page, Author Central, ads.
- You own: your website, your email list, your analytics.
- Reach: social and short video, podcasts and video, advance readers.
Author Central features and prerequisites: S13.
Your own channels. Amazon owns the store page; you own your list.
- Author website. A single page can be enough: the book, the reader promise, a sample chapter, an email sign-up and where to buy. Its job is to give search and curious readers a place to land and give you something you control.
- Email list. The one audience Amazon does not control. Offer a short free sample or related story rather than “join my newsletter.”
- Social and short video. Choose one platform you can keep up. Show the premise and cover in a few seconds. Measure what produces sign-ups, not likes.
Podcasts: guest or your own show?
FACT PodcastGuests.com states that its services are free for podcasters and that it does not charge you to be a guest or to get its free newsletter; an optional paid directory listing lets podcasters invite you directly. [S17]
FACT Spotify for Creators markets free podcast hosting. A show hosted with Spotify for Creators is made available on Spotify automatically; to reach other platforms such as Apple Podcasts you submit your show to each service yourself, using the RSS feed found in Settings > Availability. [S18] Apple requires an Apple Podcasts Connect account and a public RSS 2.0 feed. [S19] YouTube can ingest an RSS feed in some countries (verification by email code, audio-first feeds, no ads in uploaded podcast content). [S20]
YOUR DECISION For most new authors, guest appearances are the cheaper first test: direct pitches to shows your reader already hears cost time, not money. Your own show is a long commitment with slow audience growth, and makes sense mainly if you enjoy talking and have a topic wider than one book. We did not find first-party evidence that Podcastly.co offers free hosting, so we do not recommend it as a free option.
Podcast: guest or your own show?
Four quick questions. The answer is a first move, not a commitment.
Facts about directories and hosting: S17 to S20.
If you are publishing in Q4 2026 or preparing for 2027
FACT As of 26 October 2026 there are 66 days left in 2026. We do not claim that the holiday season automatically raises book sales; we found no first-party evidence for it. What follows is about timing and readiness.
What is a platform deadline, and what is only planning advice
| Item | Type | Detail |
|---|---|---|
| Pre-order manuscript deadline | Platform | At least 72 hours before release [S1] |
| Pre-order window | Platform | Up to 18 months for pre-orders created on/after 2 Sep 2026 [S1] |
| A+ Content review | Platform | Compliant content appears within eight business days [S14] |
| Typical go-live after submission | Platform estimate | Typically within 3 business days; detail page up to 72 hours [S11] |
| Manuscript freeze, cover, formatting, advance readers | BMT planning guidance | Suggested offsets in the planner below, adjustable |
Three realistic paths
1. Release in December 2026 (only if you are already close). Example date: Tuesday 15 December 2026. This works only if the manuscript is final and edited, the cover is done and the file has been tested in Kindle Previewer. Without a pre-order, upload about a week early (guidance), because KDP typically takes up to three business days and the page can take up to 72 hours. If any of those three is missing today, a compressed December launch usually means compromising quality. YOUR DECISION Waiting is a legitimate choice.
2. Pre-order now, release in January 2027. Example date: Tuesday 12 January 2027. Create the pre-order when the cover and description are final; the final file is due 72 hours before release (platform) and two weeks earlier is safer (guidance). Use the weeks before release for advance readers, the author page and your email list.
3. Prepare properly for a spring 2027 release. Example date: Tuesday 16 March 2027. The planner below shows an example: freeze about 19 January, cover about 2 February, formatted files about 16 February, advance readers about 23 February, A+ submitted by about 2 March, final file submitted about 2 March (guidance) against a hard deadline of about 13 March (platform).
Launch-date planner
Turn a release date into dated steps. Platform facts and BMT planning guidance are tagged differently; the guidance buffers are not Amazon requirements.
PLATFORM FACT comes from KDP Help (S1, S11, S14). BMT GUIDANCE is editorial planning advice with adjustable buffers.
Where preparing early genuinely helps
We do not claim a general “first-mover advantage.” These are specific mechanisms:
- A pre-order page and an author page give readers something to follow before release day. FACT Readers can follow your Author Page or a book detail page to be notified of new releases and pre-orders (depending on eligibility and their settings). [S13]
- Editing and covers book up. RULE OF THUMB Good editors and cover designers often have waiting lists. Booking early removes pressure on your freeze date.
- A+ Content takes up to eight business days and you cannot rush it. [S14]
- Moderation of ads takes up to three business days. [S16]
- Your list needs time. Readers rarely join a list in the week of a launch.
A 2027 roadmap (guidance)
2027 author launch roadmap
Three lanes, November 2026 to April 2027. Guidance, not a platform requirement. The dashed line is the pre-order file deadline for the example 16 March 2027 release.
Manuscript
- Nov 2026: Finish or revise
- Jan 2027: Freeze text
- Feb: Format and test
Platform
- Nov 2026: Author page, 1-page site
- Jan 2027: Set up pre-order
- Mar: A+ submitted
Audience
- Dec: Advance readers listed
- Feb: Advance readers read
- Mar: Launch month
- Now to end of Q4 2026: finish or revise the manuscript; define the reader; start the author page and a one-page website; list and approach advance readers; draft metadata; research podcasts that reach your reader.
- January 2027: freeze the manuscript; book editing and cover if not done; set up the pre-order when the cover and description are final.
- February to March 2027: format and test files; advance readers; A+; email warm-up; podcast outreach.
- Launch month: the 30-day plan below.
- April onward: review measurements, update the page, decide whether to start the next book.
Your first 100 readers
RULE OF THUMB This is a map, not a promise: friends and contacts who fit the reader (to read and tell others by word of mouth, not to write Amazon reviews); advance readers; a small email list; three to five communities where your reader already is (take part, do not pitch); two or three podcast guest spots; a free sample; and a store page you have tested. Record which source brought each reader. That tells you where to spend next.
First-100-readers map
Seven sources. Enter how many readers each brought and see what is left to find. A map, not a promise.
Marketing channel matrix
Ten channels compared. Sort by channel, cost, control or time to learn. Ad figures elsewhere in this guide are examples.
| Best use | Major risk | Measure | ||||
|---|---|---|---|---|---|---|
| Amazon organic (page, metadata) | Free | High | Low | Every book | Weak cover or description | Page visits, sales |
| Amazon Ads | Pay per click | Medium | Medium | After a tested page or series | Spend exceeds royalty | Spend per sale |
| Author website | Low | High | Low | Home base | No traffic by itself | Visits, sign-ups |
| Newsletter | Low | High | Medium | Launch and backlist | Slow list growth | Sign-ups, opens, sales |
| Podcast guesting | Time | Low | Medium | Reaching a new audience | Wrong show fit | Visits and sign-ups from the episode |
| YouTube / short video | Time | Medium | High | Discovery | Slow audience growth | Views to sign-ups |
| Social | Time | Low | Medium | Staying visible | Low reach | Sign-ups |
| Partnerships | Time | Medium | Medium | Cross-promotion | Mismatch | Referred readers |
| Reviews / advance readers | Free copies | Low | Medium | Optional social proof | Rule violations | Review count |
| Paid media | Money | Medium | High | Proven offers only | Burning budget | Return on spend |
The 30-day launch plan and what to measure
Book launch readiness scorecard
Twelve checks. A checklist, not a prediction of sales.
- Week -2: final proof; advance readers have the book.
- Day 0: email your list; update your website and bios; check the store page and A+ (if used).
- Weeks 1 to 4: one message a week with one reason to read; let advance readers know an optional, honest review is welcome but never required, and do not pressure anyone or ask for a particular rating; adjust description or keywords if impressions are fine but clicks are low.
What to measure. Weekly: email sign-ups, visits to your book page, KDP sales and KENP, review count, ad spend against royalties. YOUR DECISION Pick one target for your first 100 readers and one for day 30. Do not treat sales rank as the main measure; Amazon does not publish its formula.
What not to spend on (yet)
Paid trailers, large ad budgets, 3D mockups, “bestseller campaigns,” anything that promises reviews, and tools you do not yet need. Do yourself: the outline, the draft, your metadata drafts, the launch plan. Pay selectively: a developmental or copy edit, a cover and formatting for complex layouts.
Do it yourself or pay for help?
Where your time is well spent and where selective paid help is worth considering.
| Task | Approach | Why |
|---|---|---|
| Outline and draft | Do yourself | The reader journey is yours to design, and nobody else can write the daily words. |
| Metadata drafts and launch plan | Do yourself | You know the reader. Use the worksheets above. |
| Developmental edit (structure) | Pay selectively | Structure problems are the costly ones. Outside readers who are not close to you can look first. |
| Copy edit and proofread | Pay selectively | Sentences and typos are cheaper to fix than structure, so do them after structure is settled. |
| Cover | Pay if it fails the thumbnail test | Hire a designer who works in your genre. A clean simple cover beats an ambitious one. |
| Formatting | Do yourself if mostly text; pay for complex layout | Preview in Kindle Previewer either way (S3). |
| Amazon ads | Wait | Usually after a tested store page and ideally a series. Cap daily spend. |
| Trailers, 3D mockups, bestseller campaigns, review services | Skip for now | Do not pay for reviews. Amazon bars paying for or influencing them (S15). |
After you publish
Check the live page, then work the list from the 30-day plan. If you are Select-enrolled, a Free Promotion is available (up to five days per 90-day term) but earns no royalties while it runs. Refresh the description and keywords if clicks lag. Plan the next book while the first is fresh; for most authors, a second title helps readers who liked the first.
Frequently asked questions
How do I write an eBook?
Choose one reader and a one-sentence promise, validate that similar books sell, outline by the reader’s journey, draft with a daily word target, then get an outside edit.
How do I publish an eBook on Amazon?
Create the eBook in your KDP Bookshelf, upload a tested manuscript (KPF, EPUB or DOCX) and cover, complete title, description, keywords (up to seven) and categories (up to three), choose a royalty option and price, and submit. KDP says a submission typically goes live within three business days. [S3] [S5] [S6] [S11]
Do I need an ISBN for an eBook?
No. KDP says you do not need one to publish an eBook. [S6]
How much do I earn per sale?
On Amazon.com, 70% x (price – delivery) if the price is $2.99 to $12.99, with delivery at $0.15 per MB; otherwise 35% x price. Use the calculator above. [S2] [S4]
Is KDP Select worth it?
It depends on your readers and goals. It requires 90 days of Kindle exclusivity and pays Kindle+ page reads. [S8] [S10]
How far ahead can I set a pre-order?
Up to 18 months if created on or after 2 September 2026, with the final manuscript due at least 72 hours before release. [S1]
Can I use AI tools?
KDP requires disclosure of AI-generated content and says AI-assisted editing, refining and brainstorming need not be disclosed. You remain responsible for rights and compliance. [S7]
How do I get my first reviews?
If you use advance reader copies, eligible readers may choose to leave an optional, honest review. Do not require, incentivize, pressure, or influence a review or its sentiment, and do not solicit Amazon customer reviews from friends, family, or other close personal connections. [S15]
Where to go next
Pick the path that matches your stage. None of these is a sales pitch for a book you have not finished.
- I am querying agentsQuery Package Consistency Check ($99, fit-gated). It reviews the consistency of your letter, synopsis, comparable titles and sample pages and does not promise representation.
- I am not sure my manuscript is readyManuscript Readiness fit check. A short conversation about what stage your draft is at. No public price.
- I am self-publishingTell us where you are with your author page or site, and we will point you to what fits your stage.
For how readers actually find and buy books, read How Readers Find and Buy Books in 2026.
Nothing here guarantees representation, sales, rank or revenue.