Author’s Note: This is a topic near and dear to my heart. Thinking about yesterday’s post about the rise of YA fiction as a force majeure in the SF&F publishing world, it wasn’t far for me to start thinking about book length. Also, to be entirely honest, Dear Wife suggested both topics. I’ll also note: this is a very meaty (i.e. wordy) and at times contentious topic. For that reason, I am going to do something I rarely ever do on my blog: I’m implementing sectional subtitles. Why? Because this turned out to be a real, long, in-depth, even semi-scholarly article on the topic of wordcount length, with quite a bit of data and market analysis. Your conclusions will be your own, but I’ve tried to synthesize a lot of information for this article. I considered splitting the article into several posts, as I often do when a single post grows this long, but I felt that it would weaken the analysis to have the disparate elements separated onto different pages. So, instead: one long post with sectional subtitles. Finally, you’ll find I prefer the compound word “wordcount” as opposed to splitting the word into two: “word count”, which is the more common usage. The reason for this is that when I refer to “wordcount” I’m referring to a single, distinct idea: that is, the total number of words in a manuscript. Splitting the word into two diffuses this unified notion.
Books of a Certain Length
If you look around on the internet, it won’t be hard to come up with some solid advice for how long your book should be – depending on which genre and market you are writing for. I encountered advice on the issue in this post on the Magical Words blog – where you’ll find me entering the fray in the comments. There’s also this post on The Swivet. I won’t quote all the genre length guidelines these two posts suggest (which are mostly in accord). But if you’re either a fan of meaty Epic Fantasies or books like the Harry Potter series, and write in anything approaching a similar vein and genre, you might find some of these guidelines a trifle… strange. Epic Fantasy is given a high-end wordcount length suggestion of around 120,000 words. For YA it is suggested you stay under 80,000 words with some flexibility up to 100,000 in special circumstances.
For those of you unfamiliar with relative wordcount lengths, you may consider that and say to yourself: “Okay, so, what’s the big deal?”
The Challenge of a Verbose Writer
Let me first start by offering this full disclosure: my writing style tends toward the robustly wordful. For example, I’ve participated in several “Flash Fiction” challenges during the history of this blog (with most results posted here) with the goal of turning out a super-short story under 1,000 words in length. I rarely reached that goal. My first attempt at a novel, “Project SOA”, had reached the two-thirds complete mark at approximately 140,000 words before I abandoned that version of the story. I’m planning on my current novel project, “The Book of M”, to be about 125,000 words… but I fully expect it to be closer to 175,000 (based on my experience of planned length versus actual final length for other, shorter works).
Of course, I’m no professional, as yet. But I think my writing style is hitting its stride, and is reaching its maturity. I struggle to write shorter, but I usually find that the only way to say the same thing in fewer words is to make the prose so utilitarian and efficient as to lack any character and warmth. Artistically, there’s a point at which I’m not willing to compromise. When fewer words are actually better, I’d certainly prefer to use fewer words. But so far I don’t often find that to be the case: I actually find that more words are frequently called for – both in my own work, and in the work of others.
All that said, I reiterate, again, that I am not yet a professional. And perhaps there’s a reason for that. Perhaps my wordy prose is, in fact, a true sign of my amateurish lack of polished talent. Perhaps I am no true author, ready to write and polish and make my work in all ways presentable and salable – but instead a mere poseur. I do not discount this possibility. I have, after all, only one small professional publication to my name, and one semi-worthy commendation of my work. And perhaps that is all I’ll ever have.
When is Bigger Better?
But, it must be said: I believe in myself and I believe in my prose. I believe that what I write actually is quite good. I know I still have work to do get myself to the level I want to be at, and to the level I need to be at if I’m going to be successful as a writer, but I believe I’m well on that road and pointed in the right direction. And I don’t believe that my trending-somewhat-longer writing style is actually a sign of a problem in my writing style.
And I do actually believe that sometimes longer is better. Sometimes less is more, but sometimes more is more.
As I mentioned last week, I’m a fan of objective, verifiable data. And as I said above, if you are a fan of either the YA SF&F genres or the adult SF&F (especially Epic Fantasy) genres, you might find yourself bemused at these general wordcount restrictions. And why should you find yourself thusly bemused? Let’s take a look at some numbers:
A Short Survey of Wordcount Lengths of Popular YA and Adult Epic Fantasy Books
Harry Potter books by J.K. Rowling
- Harry Potter and the Sorcerer’s Stone*: 77,508
- Harry Potter and the Chamber of Secrets: 84,799
- Harry Potter and the Prisoner of Azkaban: 106,821
- Harry Potter and the Goblet of Fire: 190,858
- Harry Potter and the Order of the Phoenix: 257,154
- Harry Potter and the Half-Blood Prince: 169,441
- Harry Potter and the Deathly Hallows: 198,227
- Average: 154,973
Twilight books by Stephanie Meyer
- Twilight*: 118,975
- New Moon: 132,758
- Eclipse: 148,971
- Breaking Dawn: 186,542
- Average: 146,812
Hunger Games books by Suzanne Collins
- The Hunger Games: 99,750
- Catching Fire: 101,564
- Mockingjay: 100,269
- Average: 100,528
Middle-Earth books by J.R.R. Tolkien¹
- The Hobbit: 95,022
Dragonlance books by Margeret Weis and Tracy Hickman¹
- Dragons of Autumn Twilight*: 142,597
- Dragons of Winter Night: 130,657
- Dragons of Spring Dawning: 125,142
- Time of the Twins: 134,202
- War of the Twins: 129,237
- Test of the Twins: 96,644
- Average: 126,413
In Adult Epic Fantasy:
The Lord of the Rings by J.R.R. Tolkien
- The Fellowship of the Ring: 187,000
- The Two Towers: 155,000
- The Return of the King: 131,000
- Average: 157,677
The Wheel of Time books by Robert Jordan and Brandon Sanderson
- The Eye of the World: 305,902
- The Great Hunt: 267,078
- The Dragon Reborn: 251,392
- The Shadow Rising: 393,823
- The Fires of Heaven: 354,109
- The Lord of Chaos: 389,264
- A Crown of Swords: 295,028
- The Path of Daggers: 226,678
- Winter’s Heart: 238,789
- Crossroads of Twilight: 271,632
- Knife of Dreams: 315,163
- The Gathering Storm: 303,630
- Towers of Midnight: 325,998
- Average: 302,960
A Song of Ice and Fire books by George R.R. Martin
- A Game of Thrones: 298,000
- A Clash of Kings: 326,000
- A Storm of Swords: 424,000
- A Feast for Crows: 300,000
- A Dance with Dragons: 422,000
- Average: 354,000
Stand-alone books by Brandon Sanderson
- Elantris*: 202,675
Mistborn books by Brandon Sanderson
- The Final Empire: 214,752
- The Well of Ascension: 252,739
- The Hero of Ages: 244,201
- Average: 237,231
Stormlight Archive books by Brandon Sanderson
- The Way of Kings: 386,470
Kingkiller Chronicles books by Patrick Rothfuss
- Name of the Wind*: 255,986
- Wise Man’s Fear: 395,000
- Average: 325,493
Sword of Truth books by Terry Goodkind
- Wizard’s First Rule*: 279,840
- The Stone of Tears: 391,375
- Blood of the Fold: 239,020
- Temple of the Winds: 272,450
- Average: 295,671
Malazan Book of the Fallen by Steven Erikson
- Gardens of the Moon*: 209,000
- Deadhouse Gates: 272,000
- Memories of Ice: 358,000
- House of Chains: 306,000
- Midnight Tides: 270,000
- The Bonehunters: 365,000
- Reaper’s Gale: 386,000
- Toll the Hounds: 392,000
- Dust of Dreams: 382,000
- The Crippled God: 385,000
- Average: 332,500
What Do the Numbers Mean?
Okay, yeah, so for those of you keeping track at home: the average length of these highly popular YA books is 124,749 words. The average of the first in each YA series is 106,770 words.
Now granted, what is perhaps the most popular of the YA series here, Harry Potter, has the shortest debut novel, at 77 thousand words. Except that, actually, the first Harry Potter novel, by today’s standards, was marketed not as YA but as Middle Grades – which has an even shorter acceptable length (around 50,000 words per the linked advice above). It targeted not the Teen demographic that is the YA domain but 10-11 Year Olds. Very interesting.
It does indeed seem that children in the YA category prefer longer books.
Meanwhile in Adult Epic Fantasy Land… well… let’s start with Grandpa Tolkien, whose “Lord of the Rings” was supposed to be a single book, but was split into 3 volumes because of the limitations of print technology at the time. That would have been a single book of 473 thousand words. Even split, he sets a pace well above today’s 120,000-word-limit. The Wheel of Time gets totally crazy with wordcount (the longest series listed, by wordcount, for which I have complete data), and starts off big.
When you look at all of these popular adult Epic Fantasies together, you have an average novel length of 289,547 words – more than double the recommended length. And first books average out to a mere 259,958 words.
The common reply to an analysis like this goes as follows: “But you can’t base that on authors like these. They were already published before they started their big series. They were known quantities with proven sales records.” To which: take note of the titles marked with an asterisk. Each of these is a debut novel: not only the first in their respective series, but the author’s first published novel, ever. Almost all of them are well above the recommended wordcount limit for their respective genres. In the last decade-ish, especially, you have Steven Erikson debuting at 209 thousand, Patrick Rothfuss at 255 thousand, Brandon Sanderson at 202 thousand, and Stephanie Meyer (YA) at 119 thousand.
There are a few legitimate complaints with these figures, however. First, it cannot be said that there is a causal relationship between the quality of an author’s writing or series and the length of their books. This is correlative, but not causal: author’s whose books prove popular happen frequently to write very long books. Writing longer books won’t make your writing better or make you popular. Writing excellent-quality, highly readable and enjoyable stories will be far more important, in that regard.
Second: this analysis does not take into consideration the potential impact of any additional marketing support longer books may receive from publishers. In other words, it is possible that publishers see longer books as more significant investments, leading them to support the launch of those books with bigger marketing and publicity campaigns. In such a scenario, the success of these larger books would become a self-fulfilling prophecy. Suffice to say, however, that I have no information about the marketing spend by any given publisher on any given title. Data like that is proprietary, not public.
Third: my survey of popular authors, books, and series is not in any way scientific. I took the most popular books I could think of in these two categories. There may be others that I didn’t think of. None of these is reflective of actual sales of these books. A truly scientific survey would consider actual sales numbers of the books, and would ask a random sampling of readers questions about their favorite books, preferred genres, favorite authors, and what they liked and disliked about them (with some carefully-worded questions probing each respondent’s reaction to book length). I’m not in any position to perform such a scientific survey, so you get this mostly anecdotal survey instead. But even so, this is some pretty powerful anecdotal evidence.
However, with that all said, caution is still in order. This list includes a number of exceptional and/or exceptionally-well-received books. (I myself have not read all that I would like to read, alas. You will note the presence of a fair number of them on my current to-read list.) Those of us who are “undiscovered authors”, however, have not proven ourselves yet to be similarly exceptional. Some of us will succeed in publishing and never achieve this exceptional level of sales. Others of us will never succeed in publishing at all. It’s the rare bird who will achieve that vaunted status. And though clearly the publishers will bend or break this wordcount rule, by all accounts they do not do so lightly.
The Case for Shorter Books
So why exactly do they have this rule, in the first place? Both of the linked pages providing this advice cite as the reason for the rule: book sellers. There is limited space for books on the declining number of book seller shelves. And book sellers, wanting to maximize the value of their shelf space, naturally want to pack as many books onto those shelves as they can fit. And as the number of outlets for books declines, the negotiating power of the remaining sales channels increases. (This is the opposite of what we commonly call a monopoly: a monopoly happens when only one seller produces a product, although colloquially we use the term to refer to occasions when there are very, very few sellers, because the effect is largely the same. The situation in book sales, with the decline of many mom-n-pop shops has become a monopsony, where one (or very few) buyers are buying books from book publishers, which gives those buyers increased leverage over the seller (the publishers). Of course, the reality is a bit more complicated than that.) The net effect: fewer book sellers demanding changes of the publishers, and the publishers accommodating. And besides that… it’s cheaper and easier to bind a smaller book (smaller books can use cheaper binding techniques), which introduces a cost-constraint. So there are some very good reasons for publishers to prefer shorter books – and ignoring this fact isn’t likely immediately to endear a brand new author to said publishers, unless that publisher is convinced that the new author’s story is worth it.
And besides all that, it is almost certainly the case that some readers in fact do prefer shorter books. Not all readers are created equal, and their preferences exist on a multidimensional spectrum. While on the one hand many of the bestsellers of the epic fantasy and YA SF&F genres are also among the biggest doorstoppers in the market, those books are frequently subjected to the criticism that they’re too long. Still, it’s hard to make the case that this observation is objectively true when these books sell so many copies. For a few readers, it undoubtedly is true. For the majority of readers? Their purchasing and reading habits seem to speak for them.
What does that mean? You have to be more than just good. You have to be among the best.
Still… the evidence, anecdotal or not, above gives me reason to pause and consider: is following the recommended course of action wise?
Thinking like a Businessman: Market Conditions and Trends Affecting Epic Fantasy and YA SF&F
Earlier I linked to the article I wrote last week about sales data. Look, I just did it again! In that article, I asserted that a writer needs to treat their work like a business. I really believe that. As a writer, you are self-employed. You produce a product, and then you try to sell that product to other people. In that regard, what are the business implications of all this? Well, there are two sides to the coin. First, there is the traditional market. If you are going to pursue a sales channel model that moves through the traditional publishing market, considering the constraints of the traditional industry might very well be the wisest course of action. It might be better,the data suggests, would be to write the most amazing epic ever – no, really, it’s got to be pretty amazing and publishers have to agree – and to ignore the rules on wordcount and manuscript length. But you’re going to have a hard time convincing a publisher that your work rises to that level. Few enough books get published, fewer still will qualify for the selective enforcement of these novel length guidelines. Or, perhaps, the best thing to do is to follow those guidelines faithfully and prove your mettle to a publisher with shorter books in order to make room for yourself to grow into a career writing longer books. (Several of the authors on my list above didn’t start out writing mega-long epics, but shorter works instead.)
But there’s another side to this coin: the traditional publishing world is even as we speak being disrupted by new technologies and sales channels and disintermediation. And that suggests something very different about what might be the wisest course of action. Without more statistically accurate data on the tastes and preferences of consumers, a little digging into the lengths of the most popular books in your genre, like what I’ve done here, is probably the best you’re going to be able to do to find out the length that actual consumers prefer. That, and ask as many of those readers as you can about it. So this data may not be the best indicator, but what it does tell us is that readers like longer books. In the disintermediated market, then, the best thing to do might actually be to write longer books – because it appears that’s what readers prefer. (There is probably an upper-bound to this, and an ideal length. The data collected suggests to me that the ideal length is somewhere in the 200 to 275 thousand word range for epic fantasy, but that’s a shot in the dark, really.)
Even without the disintermediating effects of digital self-publication, increasingly more and more book sales are through digital e-readers. This will prove to be true even for books sold by traditional publishers. E-books are not constrained in the same way that physical books are. (Digital storage can be an issue, but books of any length compress to generally very small files.) This fact negates the physical stocking limitations of traditional book sellers and the binding constraints of traditional publishers. Inasmuch as at some future point the majority of books sold will be in a digital format, nearly all of the assumptions that make the case for shorter books today go out the window. When that day comes, if my thesis is correct and readers generally prefer books that are considerably longer than publishers today are typically willing to consider, then there will be significant pressures on traditional publishers to relax those artificial limitations and to consider longer books. (Even those books that are merely very good and not necessarily epicly amazing.)
I don’t know when that tipping point is going to occur. But I am very confident that it will occur, and that at some near-future point publishers and agents both will begin to relax this wordcount limitation. I believe it will happen because I believe it has to happen for traditional publishers to remain relevant in the age of disintermediation. Intransigence isn’t a viable business strategy. Ultimately, publishers are serving the market needs of a group of customers, and if those customers want longer books, when the cost constraints on supplying that need evaporate or vanish, there won’t be any reason for the publishers not to meet that market need. Indeed, it will be more profitable for publishers to find that sweet spot – the ideal book length – than for them to ignore it because of old and passed business constraints.
What does this mean for me? Or for you, Dear Reader?
Honestly, it’s hard to say. As I said, I don’t know when the shift in the traditional market will occur. It almost certainly will occur, but the future is murky and uncertain. It could be next year, it could be five years, it could be ten. (If it takes much longer than that, I would be surprised; if it doesn’t happen, I might have to reassess my conclusion that readers like longer books.) For some of us, we’ll need much of that time to hone our craft as it is, so this doesn’t mean much of anything to us. Those writers will write what they want to write when they’ve reached a publishable skill level, and if they’re really any good, they will find a market for their work. Some of us are a little closer to being ready to enter the market. They have books ready now, or will have books ready in the next year or two, that are polished and publishable. These writers are in a trickier spot – they have to decide whether to write to traditional publishing’s length constraints, or maybe to ignore those constraints and self-publish. But that’s a decision authors today have to make for themselves anyway, so I don’t think that your debut novel’s length will be the critical, deciding factor in that decision.
For myself? I must first ask myself: is my writing any good? If it’s good, then the only thing I want to worry about is how long does the story have to be to tell the story the way I want to tell it? I want my prose polished and beautiful and as tight and clean as I can make it. But I have no desire to sacrifice part of my story to fit into any artificial length constraints. Were I trying to send my book out now, today, that might be a problem for me. But I’m still at the early stages of this novel. Tomorrow is another day. Today, I write. Tomorrow, we’ll see if those old rules are still around, and if they are, I’ll decide what to do then. If they’ve gone the way of all the earth, well. You shall not find me weeping by their graveside.
What do you think, good readers and writers all: how long are your favorite novels? How long would you prefer the books you read to be? How long is the novel you’re working on now? How long do you think it should be, everything else being equal? Sound off with your thoughts on wordcounts and book length.
Wordcount Figures in italics are estimated numbers; for LotR, ASoIaF, and Malazan they are rounded from other sources, in the links below, and for others (i.e. Sword of Truth) they are by my own calculations. My methodology was to count words per line on several pages (about 3 pages worth using either Amazon’s “Look Inside” feature to get the first few pages, or “Google Books” to find random pages) to generate an average number of words per line, multiplying this first figure by the number of lines per page (which is almost always 40, it seems) and multiplying that by the number of pages (per the same source that generated the copy used for the above calculations). I discounted this somewhat because first pages of chapters and last pages of chapters typically have fewer lines of text. I gave up on this after doing two “Sword of Truth” books – thus the short list of titles there – because of how time consuming it was and never attempted another series I was going to include in my analysis, “Memory, Sorrow, & Thorn”. I actually did this for several books besides the two Sword of Truth books for which I used estimated figures before I found more accurate figures reported elsewhere online.
*Titles marked with an asterisk are books that, as far as I am aware, are the author’s debut, first-ever-published novel.
¹Both The Hobbit and the Dragonlance books were published before the modern “YA” designation came into being. But The Hobbit was marketed as a children’s book, and would almost certainly qualify today as “YA”. The Dragonlance books, meanwhile, I believe are also best read in one’s teenage years. I loved them when I was younger, but I daresay were I to encounter them for the first time today, as an adult, I would find them less to my tastes. These books appear to have approximately a sixth-grade reading level, which is just about the age I first discovered them. For those reasons, I’ve classified them here as “YA”.
(Note: A fair number of books, but not all, have wordcounts listed on Amazon: near the bottom of the page for a book, below the reviews, is a section with the heading “Inside This Book”. Some books, in this section, will include a link that reads “Text Stats” in which you’ll find the wordcount. Relatively few, that I found, have this new feature.)