Reviews are the one asset in a store listing that you cannot make, buy or design. They accumulate slowly, they decide whether a stranger clicks, and almost every self-published book has too few of them for the same reason: nobody asked.
Why they matter more than the number suggests
Not because readers read them. Most do not read past the first line. They matter because of what a count signals and what a store does with it.
- A book with two reviews looks unfinished. The count is read as evidence that the book exists in the world, before anyone reads a word of any review.
- Stores use them as a ranking input, alongside sales and click-through. A book that converts browsers into buyers gets shown to more browsers.
- They are the only social proof on the page you did not write yourself. The description is yours, the cover is yours, the reviews are not — which is exactly why they carry weight.
- Promotional services require them. Most newsletter promotions have a minimum review count and rating before they will take your money.
What every store forbids
Read this before the tactics, because the penalty is the listing rather than a warning. Terms differ in detail and agree completely in substance.
| Not allowed | Why it is detected | Consequence |
|---|---|---|
| Paying for reviews | Payment trails and reviewer patterns | Reviews removed, listing at risk |
| Review swaps with other authors | Reciprocal patterns between accounts | Both accounts penalised |
| Reviews from family or household | Linked accounts and addresses | Removed silently |
| Offering anything in exchange | Reported by the reader you asked | Account review |
| Reviewing your own book | Obvious | Account closure |
| Asking for a specific rating | Reported, and it is in the terms | Removed |
The line worth internalising: you may ask for a review, and you may not ask for a good review, or offer anything at all in return. A free copy given before publication is fine and standard; a free copy given on condition of a review is not.
Where the first twenty come from
Almost entirely from people you contacted by name, before the book was on sale. There is no version of this where a stranger finds an unreviewed book and reviews it.
- Early readers. Ten to twenty people asked individually, six weeks out, given the EPUB and the PDF. This is the single largest source and it is covered as a launch job in the book launch checklist.
- Your email list. The people who bought first. One email, two weeks after purchase, asking once.
- Anyone who has told you the book helped. A reply, a DM, a comment. Those people will review it and have simply not thought of it.
- Professional contacts in the subject. For non-fiction, colleagues who would have opinions anyway.
- Book bloggers and reviewers in the genre, for fiction. A slow channel with a low hit rate, worth starting early and not worth counting on.
The page at the back of the book
This is the highest-yielding thing on the list and the one most books do not have. Every reader who finishes reaches it, and they reach it at the only moment they will ever feel strongly about the book.
Keep it short, put it immediately after the last line of the last chapter rather than after three appendices, and make it a link rather than an instruction to go and find the listing. In an EPUB that link is tappable; in a PDF it is clickable; in print it is a short URL or a QR code.
## Before you go
If this was useful, a short review helps more than
you would think — it is the main way other people
decide whether to read it.
[Leave a review](https://your-store-link)
It takes a minute, and one sentence is plenty.Three things that make this work: "one sentence is plenty" removes the reason most people do not (they think a review has to be an essay), the link goes directly to the review form rather than the product page, and there is no mention of what the review should say. Front and back matter covers where this page sits among the rest of the furniture.
The ask that works
Short, specific, no pressure, and one request only. The most common failure is an email that asks for a review, a share, a mention and a reply, and therefore gets none of them.
| Instead of | Write | Why |
|---|---|---|
| "Please leave a 5-star review" | "If you have a minute, a review helps" | The first breaks store terms |
| "Let me know what you thought!" | "One sentence is plenty" | Removes the size of the job |
| A paragraph of context | Two sentences and a link | Read on a phone, in four seconds |
| "Reviews are so important to authors" | "It is how other people decide whether to read it" | A reason about the reader, not about you |
| Asking three times | Asking once, well | The second ask costs you the relationship |
The timing that gets answered
- Early readers: on publication day. Reviews landing on day one do more than the same reviews landing in week three, and most early readers need reminding that the day has arrived.
- Buyers: about two weeks after purchase. Long enough to have read it, short enough to remember it.
- Never before they have read it. A request attached to the download reads as a condition of sale.
- Once. A second reminder converts a small number of people and annoys the rest.
The two-week email is the one that requires infrastructure rather than effort, and if you sell direct you already have it — how to deliver an ebook to buyers covers the delivery sequence that this email belongs at the end of. Store sales give you no email address at all, which is one more entry on the long list of reasons to sell direct as well: where to sell your ebook.
What to do about a bad one
Nothing, publicly. Ever. Replying to a critical review is the single most reliable way to turn one reader's opinion into a story about the author, and it is a story that outlives the book.
- Do not reply, anywhere. Not on the listing, not on social media, not obliquely.
- Read it once for information. Three reviews saying the middle drags is a structural note you can act on.
- Fix what is fixable. A typo cluster or a confusing chapter is a re-upload, which is the advantage a digital book has over a printed one.
- Report only genuine abuse — a review that is about you rather than the book, or one that reveals a competitor. Rarely worth the effort.
- Expect a spread. A book with nothing but five-star reviews reads as suspicious. A few threes make the rest believable.
A realistic picture
Roughly one review per one hundred copies sold is a normal, unremarkable rate for a book that does not ask. Asking well moves it, and it is still a small number — twenty reviews is a genuine achievement for a first self-published book, not a disappointing start.
Which sets the expectation properly: this is a slow compounding asset, not a launch tactic. The back page runs for the life of the book, the two-week email runs for every buyer forever, and both were set up once. Everything else on this list is a week of work that produces the first twenty and then stops.