"How long should my novel be?" is the question that gets asked in every writing group on the internet, usually by someone who's just hit 140,000 words on a debut romantic suspense and is starting to suspect they may have overdone it. They have. But the answer is more useful than just "shorter."
Word count matters in indie publishing for three separate reasons, and they don't all pull in the same direction. Print costs go up with page count. KU earnings go up with page reads. Reader expectations are set by whatever's already on the shelf. Get the balance wrong on a debut and you've made a book that's expensive to print, awkward to price, and quietly out of step with what readers in that genre actually buy.
Here's what the numbers look like, and what to do with them.
Why word count actually matters
For trad publishing, word count matters because agents and editors use it as an early filter. A 200,000-word debut from an unknown author goes into the no pile before anyone reads page one — not because it's bad, but because it's going to be expensive to produce and a hard sell to bookshops.
Indie authors don't have that filter. We can publish a 250,000-word debut if we want to. The market, though, will quietly apply its own version of the same rule.
The reasons it matters for us are slightly different:
- Print cost. KDP charges per page for paperbacks and hardcovers. A 100,000-word book runs to roughly 350-400 pages depending on trim size and font. A 200,000-word book is closer to 700. That doubles your per-unit print cost and forces your cover price up — sometimes past the point where readers will buy.
- KENP earnings. If you're in Kindle Unlimited, every page a reader gets through pays you something. At the time of writing, that's around half a US cent per Kindle Edition Normalized Page — so a longer book read to completion pays more than a shorter one. The catch is that readers are far less likely to finish a long book in the first place.
- Reader expectations. Every genre has an unspoken contract about length. Romance readers expect a certain number of hours. Epic fantasy readers expect to disappear for a week. Cosy mystery readers expect to finish on the train home. Break the contract and you'll see it in the reviews.
- Sample length. The Amazon "Look Inside" preview shows roughly the first 10% of your book. A 50,000-word novel gives readers a 5,000-word sample. A 200,000-word novel gives them 20,000. That's not always a good thing — it changes how decisive the opening has to be.
The numbers, by genre
What follows is the range readers and reviewers expect. Below the bottom number, readers tend to feel short-changed. Above the top, they tend to drift away. Neither is fatal, but both make life harder than it needs to be.
Adult literary and general fiction
- Literary fiction: 80,000 – 110,000
- General/upmarket fiction: 80,000 – 100,000
- Historical fiction: 100,000 – 150,000 (some sweep into 180,000+ for multi-generational sagas)
Crime, thriller, mystery
- Thriller / suspense: 70,000 – 95,000
- Police procedural: 80,000 – 100,000
- Cosy mystery: 60,000 – 80,000
- Legal / psychological thriller: 80,000 – 100,000
Romance
The most varied of all the genres, because the sub-categories pull in different directions.
- Category romance (Harlequin/Mills & Boon line books): 50,000 – 55,000, often to the word.
- Contemporary romance: 60,000 – 90,000
- Romantic suspense: 80,000 – 100,000
- Historical romance: 90,000 – 110,000
- Romantasy: 100,000 – 150,000 (Sarah J. Maas territory has skewed expectations upward)
Science fiction and fantasy
- Science fiction: 90,000 – 125,000
- Urban fantasy / paranormal: 70,000 – 90,000
- Epic / high fantasy: 100,000 – 200,000 (Brandon Sanderson regularly clears 400,000, but he is not a debut indie)
- Space opera: 90,000 – 150,000
Young Adult and Middle Grade
- YA contemporary: 50,000 – 80,000
- YA fantasy / sci-fi: 70,000 – 100,000
- Middle Grade: 25,000 – 50,000
- Early chapter books: 5,000 – 15,000
Shorter forms
- Novella: 17,500 – 40,000
- Novelette: 7,500 – 17,500
- Short story: under 7,500
- Flash fiction: under 1,000
These aren't ironclad rules. Plenty of genre-defining books have ignored them. Gone Girl is over 145,000 words. The Road is barely 60,000. The point isn't that the numbers can't be broken — it's that breaking them is a decision you should make on purpose, not by accident.
The KU calculus
If you're in Kindle Unlimited, the maths gets more interesting. Amazon pays you per page read, not per borrow. So in theory, a longer book pays more — assuming readers actually finish it.
They mostly don't. Completion rates drop steeply with length. A 60,000-word book might see 70% of borrowers reach the end. A 150,000-word book often sees 30% or less. The earnings curve flattens hard somewhere around 100,000 words, because the extra pages are being skipped.
The other side of this is that short books — under about 30,000 words — get a smaller share of the KU pot than they probably deserve, and they're often discoverable only to readers who specifically want a quick read. The KU sweet spot for most genres ends up being 70,000 to 100,000 words. Long enough to feel like a proper novel, short enough that people finish it.
The print cost calculus
For paperback editions, the per-page cost is where length bites. As a rough guide on KDP at the time of writing, a black-and-white paperback in the UK runs to roughly £0.85 plus around £0.012 per page. A 350-page book costs you about £5 to print. A 700-page book costs around £9.20.
You can't easily price a 700-page paperback under £14.99 without losing money. You can comfortably price a 350-page paperback at £9.99 and still take a couple of quid in royalty. The longer book asks readers to pay more for something they're less likely to finish.
This is why "as long as it needs to be" — the standard writing-class answer — falls apart slightly under the indie cosh. The book really does need to be as long as the story demands. It also needs to be a length the market will bear. Most of the time those numbers are close. Sometimes they're not, and that's a tension worth recognising before you publish, not after.
Common indie length mistakes
Three patterns crop up repeatedly:
- The 200,000-word debut. Almost always either two books pretending to be one, or a first draft that hasn't been cut yet. Splitting in half (and adding new endings to both) usually produces a stronger pair than the one fat brick.
- The 40,000-word "novel." If you're selling it as a novel, readers expect a novel. Marketing a novella as a novel earns you a one-star review the same week it goes live. Marketing it honestly as a novella, on the other hand, is fine — there's a real audience for shorter reads.
- The genre-mismatched length. A 60,000-word epic fantasy reads as undercooked. A 130,000-word cosy mystery reads as overcooked. Genre conventions exist because readers internalised them. Working against them costs you the readers who came in expecting the standard.
What to actually do
Write the book first. Worry about length last. If you start trying to hit a number, the prose suffers and you end up either padding the middle or amputating the end.
Once a draft is done, check it against the ranges above. If you're inside the band, leave it alone. If you're noticeably outside, two questions are worth asking. First: am I outside it because the story demands it, or because I haven't cut what doesn't earn its place? Second: is this length going to make my book harder to sell, harder to price, or harder for readers to finish?
If the honest answers are "I haven't cut enough" and "yes," the editing pass is where you fix it. If the honest answers are "the story demands it" and "I've thought about it" — then it's the right length and the rest is your problem to solve in cover, blurb, and pricing.
The honest summary
The ranges in this post are descriptive, not prescriptive. They describe what's already on the shelves in each genre, because that's what readers in each genre have come to expect. You can break them. Some of the most interesting indie books do. But the more you break them, the more work the rest of your launch has to do to compensate — your cover, your blurb, your categories, your price, your marketing, all of it has to explain why this book is shaped differently from the ones beside it.
The boring infrastructure of a launch — the bits where you decide imprint, ISBN, format, price, and yes, length — is where most launches are quietly made or unmade. That's the bit I built WIPsage to help with. Have a look if you'd like a clearer view of what order to take those decisions in.