A launch is not an event, it is a deadline that forces a list of jobs to be finished. Almost nothing on that list can be done in launch week, which is why launch week goes badly for people who start planning it in launch week.
Eight weeks out: the book itself
Everything here is about the object. None of it is marketing, and all of it constrains the marketing that comes later.
- Final manuscript, proofread by someone else. Not spellcheck. A person, reading it once, from the beginning.
- Both files exported and opened. A PDF for direct buyers and print, a reflowable EPUB for stores — PDF or EPUB covers why the answer is both, and Markdown to EPUB is the export.
- Front and back matter in place. Title page, copyright, contents, and a last page that tells the reader what to do next: front and back matter.
- Cover finished, checked at thumbnail size. It will be seen at 100 pixels wide before it is seen at any other size — ebook cover dimensions.
- EPUBCheck clean, so a store rejection is not the thing that moves your date: EPUB validation errors.
Eight weeks out: the words that sell it
Written once, used in every store, on your own site, in every email and in every post for the next two years. This is the highest-leverage hour in the whole list.
- The description. Two hundred words that earn the click — how to write a book description.
- Title, subtitle, author name exactly as they will appear, spelled identically everywhere.
- Two categories and seven keywords, chosen against the competition you can realistically place among: book metadata.
- A one-sentence version for a bio line, a reply, a conference programme.
- The price, decided before you see a store's suggestion — how to price an ebook.
Six weeks out: early readers
The single most useful thing you can do that is not writing. A dozen people who have read the book before launch produce the reviews that make the first fifty sales possible, and they find the mistakes your proofreader did not.
- Ask ten to twenty people directly. Individually, by name. A public call produces volunteers who do not read it.
- Send the EPUB and the PDF, and say which you would prefer they used.
- Ask for one specific thing — where they stopped reading, or what they expected that was not there. "Any feedback" produces "I liked it".
- Tell them the date and that a review on the day matters more than a review later. It does, and most people do not know that.
- Never pay for reviews or trade them. Every store treats this as fraud, and the penalty is the listing.
Four weeks out: where people can buy it
| Set up | Why now | Time |
|---|---|---|
| Direct storefront | The highest margin, and the only place you get the email address | An afternoon |
| Amazon KDP listing | Where strangers find it | An hour, plus review time |
| Apple, Kobo, Google Play | Directly or through an aggregator | An hour each |
| Delivery and download | The thing that breaks on launch day | Test it twice |
| A page on your own site | Somewhere to link that you control | An hour |
Where to sell your ebook has the arithmetic of each channel; the launch-specific point is that they take days to review a file, so a listing submitted in launch week is not live in launch week. And test the buying flow yourself, with a real card, on a phone — how to deliver an ebook to buyers is the failure that ruins an otherwise good day.
Two weeks out: tell people it exists
The mistake is announcing a book once, on the day. Announce it repeatedly, before the day, in ways that are not announcements.
- Post the work, not the news. A paragraph from a chapter, a diagram, the thing you got wrong in the first draft. People share the content and remember the book.
- Email your list twice before launch — a heads-up with a date, then a preview of something useful from inside it.
- Give away one real chapter as a PDF, with a link to buy at the end: the lead magnet PDF.
- Line up anywhere that would mention it — a newsletter, a podcast, a professional body. All of them plan weeks ahead, which is why this is not a launch-week job.
- Set a pre-order only if you will make the date. A missed pre-order is a penalty on some stores and a broken promise on all of them.
Launch week
- Publish on a Tuesday or Wednesday. Not Friday, when everyone stops reading email, and not Monday, when they delete it.
- Email the list first, before any public post. They are the people most likely to buy, and they should not learn it from a feed.
- Ask the early readers to post their reviews today. Politely, individually, and once.
- Buy your own copy. It is the only way to see what the buyer sees, and it catches the broken link.
- Answer everything. Every reply, every comment, all week. Launch week attention is the only free attention you get.
- Do not check the dashboard hourly. Reporting lags, the numbers will look wrong, and you will make a bad decision on bad data.
The month after, which is where the money is
- Fix what the reviews found. A typo cluster or a confusing chapter is a re-upload, not a second edition. This is the advantage you have over a printed book — use it.
- Keep asking for reviews. A line at the end of the book, and an email a fortnight after purchase. Reviews are the compounding asset in every store.
- Bundle it. The book plus the templates, the worksheets, the checklist — the same words at three times the price on your own site.
- Write about the subject, not the book. Every piece that ranks is a permanent route to the listing, which is the only marketing that keeps working while you sleep.
- Leave the price alone for ninety days. Then change one thing, and only one.
And keep the manuscript in a form you can edit. The reason a plain-text source matters at launch is not tidiness — it is that fixing chapter nine and re-exporting both files is twenty minutes rather than an afternoon in a layout application, and a book you can cheaply correct is one you will actually correct.