THE BOOK IS THE PROOF. THE SITE SHOULD SELL IT.

Amazon sells this book. A website sells you, which is what sells the next one. The retailer keeps the reader's email address and their card details and gives you a sales rank, so every reader you earn through them is a reader you have to buy again on launch day. The people who leave an email address on your own site are the only audience you own outright, and for most authors that list is worth more than any single campaign.

Five things. The book, with the cover, one sentence on what it does for the reader, and a buy button, all above the fold. A way to leave an email address, with one clear reason to. A page about you written for a reader rather than a bio in the third person. A press kit. And, if you speak, a page that answers what you talk about, to whom and how to book you. Everything else is optional and most author sites have the optional parts and not these.

Before, and earlier than feels necessary. The site is where pre-launch attention goes to be captured, and a list built in the three months before publication is what makes launch week work. If the book is already out, build it now anyway: the reader who finished it last night is looking for you tonight. A site that goes live a fortnight after launch missed the only week the whole industry is pointed at you.

Almost never. One site with a page per book, and the newest one on the front, keeps every bit of authority and every link you have ever earned in one place. A dedicated domain per title splits that and leaves you with a site to maintain for a book that stopped selling. The exception is a title with a genuinely separate audience under a pen name, where mixing the two would confuse both.

Give the reader the choice, and put their likely one first. A single Amazon link loses everybody outside its home country, everybody who reads on Kobo, everybody who wants the audiobook and everybody who would rather order through a bookshop. A short list of retailers converts better than a single link, and if you sell a signed copy or a direct edition yourself, that goes at the top, because it is the only version where you keep the margin and the customer.

Ask once, in the same place on every page, and offer something specific. A pop-up that arrives four seconds after somebody lands is why most readers now reflexively close them. What works is a fixed sign-up block with a real reason attached: the first chapter, a worked example from the book, the notes you cut. One offer, one field, and no second page asking their company size.

Everything a producer or a journalist would otherwise have to email you for, downloadable, at the sizes they need: a short and a long biography, two headshots, the cover at print resolution, the blurb, your topics, previous appearances and one contact address. It sounds administrative and it is the highest-converting page on many author sites, because the person reading it has already decided to feature you and is looking for a reason not to have to wait.

The structure is the same and the emphasis moves. The series and the reading order matter more than a press kit, the list is usually built on the promise of the next release rather than a lead magnet, and the design carries genre signals that tell a browsing reader in one second whether this is for them. What does not change is that the retailer keeps your reader unless your own site gets there first.

It is not, and it is worth understanding why rather than resenting them. That page exists to sell one title during one campaign, it lives on their domain, and it will be quietly retired when the campaign is. You cannot build a list on it, you do not control it, and it does not follow you to your next publisher. Have both. Theirs is a marketing asset for the book and yours is the asset for your career.