Guide
Making money from ebooks when nobody knows who you are
The median self-published book sells in the low tens of copies. That is not a reason not to write one. It is a reason to be deliberate about which book you write and what you expect it to do for you.
Short answer
A first ebook with no audience behind it typically sells in the low tens of copies, so treat royalties as the smallest of the three ways it can pay. The reliable routes are a narrow subject with a crowd already attached, a book that leads to paid work you already do, and a back catalogue where book four sells books one to three. Pick the subject for its audience, not for your interest in it.
Start from the honest number
Self-publishing sales follow a power law. A very small number of books sell extremely well and pull the average far above what almost anyone experiences. The median book sells in the low tens of copies across its entire life.
At $2.99 with a 70% royalty, twenty copies is forty-two dollars. No pricing decision, cover tweak or keyword fixes that, because the constraint is not conversion. It is that nobody with the problem your book solves has ever seen it.
Which means the useful question is not "how do I sell more copies" but "which of the three ways a book can pay am I actually going for".
Route one: a subject with a crowd already attached
The single highest-leverage decision is the subject, and it is made before a word is written. A book competes for attention with every other book in its category. Narrow the category until there is almost nothing in it.
| Competing with everyone | Competing with nobody |
|---|---|
| A book about productivity | Scheduling for a single-chair barbershop |
| A guide to investing | What to do with a UK pension when you move to Portugal |
| Learning to cook | Cooking for one after forty years of cooking for four |
| A parenting book | The first year of a deaf child in a hearing family |
The right-hand column looks smaller and earns more, for three reasons. Those readers are searching a phrase almost nobody has written for. They recognise themselves instantly, which is most of the buying decision. And they tell each other, because the people in each of those situations already talk to each other somewhere.
The test: can you name the forum, group or subreddit where your reader already is? If you cannot, the subject is still too broad.
Route two: the book as a front door
For anyone who sells a service, the royalties are the smallest part of what a book is worth. A short, genuinely useful book does something no amount of advertising does: it puts you in someone's hands for two hours while they decide whether you know what you are talking about.
- A consultant whose book brings two clients a year has earned more from it than almost any novelist earns from one.
- A coach, therapist or accountant gets a credential that outlasts every post they will ever write.
- A software business gets a guide that ranks for the problem their product solves.
Books written for this purpose should be short — 15,000 to 25,000 words — and should solve the reader's problem completely, not tease at it. A book that withholds the answer to sell the service reads as a brochure and converts worse than one that simply helps.
Price it low or give it away. The mailing list you build is the asset; the royalties are rounding.
Route three: the back catalogue
Almost nobody makes money from one book. The pattern that works is a shelf: each new release brings readers who then buy the earlier ones, and each book keeps earning quietly for years.
This only compounds if the books belong together. Four books for the same reader, ideally in a visible series, sell each other. Four books on unrelated subjects are four separate failures. Decide who your reader is and write the next thing for the same person.
The practical consequence: a second book is worth far more than doubling the marketing on the first.
What actually moves copies with no audience
- The cover. Readers browse thumbnails. A cover that does not say its genre in one glance has already lost, whatever is inside.
- The subtitle. For non-fiction this is the whole search strategy and the whole pitch. Put the reader and the outcome in it.
- The first page. Everyone reads the sample. If page one is throat-clearing, nothing else you did matters.
- Being where your reader already is. Not "marketing" — being a useful member of the one place those people gather, for months, before and after the book exists.
- The next book. More reliable than anything on this list.
A realistic first year
Write a narrow book for a findable group. Expect little for six months. Publish a second one for the same reader. Keep the buyers who come to you directly, because that list is the only asset that compounds. Judge the project on the fourth book, not the first.
Anyone promising a different shape to this is selling a course.
Neubook writes the whole book free and charges once for the files, so testing a narrow subject costs an evening rather than a year.
Write a book, freeSources
- Amazon KDP royalty rates
- KDP Select terms and the exclusivity requirement
- Payhip pricing
- Power law distributions, which is what book sales are
- The long tail, and why narrow subjects work
Common questions
How many copies will my first book sell?
Without an audience, most likely between zero and fifty in its first year. Book sales follow a power law: a small number sell very well and the median sells almost nothing. Plan on the median and be pleased to be wrong.
What kind of ebook sells best with no following?
Narrow, practical non-fiction for a group that already congregates somewhere. A guide for people who run a single-chair barbershop will out-earn a general business book, because the audience recognises itself and there is nothing else for them.
Can an ebook make money without being sold?
Yes, and this is often the larger number. A free book that brings you consulting clients, students or customers can be worth far more than its royalties. For many professionals the book is a business card that people keep.
How long until it earns anything?
Months, usually. Search visibility, reviews and word of mouth all compound slowly. Books that sell nothing for six months and then sell steadily for years are completely normal, and books that spike at launch and die are also normal.
Is it worth writing more than one?
Almost always. The economics of self-publishing are back-catalogue economics: each new book sells the previous ones. One book is an experiment; four is a business.