How to spot a bestseller?

[Brian G Turner]

At article at NPR suggests that publishing houses are trying to collect data to provide a basis for determining if a book will sell well, or not:

Publishers' Dilemma: Judge A Book By Its Data Or Trust The Editor's Gut?

I have to admit at being surprised - my personal opinion, at present, is that is a book is structurally well-developed then it should prove to be especially engaging. By that I mean clearly show settings, stakes, and especially conflict, early on.

However, I get the impression that a lot of editors are focused on the quality of the prose.

So while some people will complain that the prose in Stephanie Meyer's Twilight is poor, I'd suggest it's a structurally accomplished book and therefore no surprise it was successful (though perhaps not so successful).

Interestingly, there's also a piece I came across at the same time about how publishers failed to spot the appeal of the Harry Potter stories:

Why were so many publishers unable to see the massive potential for Harry Potter when they first read it? - Quora

There are some interesting points suggested, though IIRC the first book originally started with a prologue used toward the end of the series.

Even still, it is a structurally well-developed story, with a superb use of conflict to generate sympathy for Harry under the Dursley family.

This is also a reason why I was really surprised that the beancounters at a Big 6 publisher turned down Jo Zebedee's Inish Carraig - IMO it is a structurally accomplished novel that has much in common with modern best sellers.

 

[Jo Zebedee]

I think structure is only part of the equation (and, Ty, btw - although I think Inish could have been stronger in the middle section and that Sunset over Abendau is structured much better, partly because of lessons learned from IC). Sure, it gets pages turning but without satisfying prose it becomes boring and without engaging characters is worthless. So, for me, only one part of an equation.

 

[Stephen Palmer]

Structure is only a small part imo. It's all about empathy of the reader with the characters, readability (narrative drive), and more…

Plus, when an author gets to a certain level of popularity such considerations fail to have any meaning. I mean, look at the inexplicable popularity of Terry Brookes…

 

[Overread]

There is certainly a threshold for technical quality in anything that allows the thing to progress from small time to big time. However that threshold is directly proportional to the market its being marketed toward. So certainly there will be hard data in there which can be understood. Indeed for mass market appeal; even if dealing only with one country; there will be patterns such as the style and use of language; the technical vs descriptive terms; the pace and flow of the book etc...


I think also though we have to consider time and market content as a factor; sometimes something becomes popular not because of how but what is written. Harry Potter is a prime example; the writing is good but its not anything special nor outstanding. However it was a well written story about something that managed to tap into something that wasn't there in the market.

Then we have to consider marketing because that's a big thing. Look at how Game of Thrones went from popular fantasy to mainstream purely by the huge marketing that went on with regard to the TV series. If we look at other markets like PC games many top titles like Call of Duty (COD) are often where they are because the company behind it put millions into marketing and indeed more than they spent on the game itself. Marketing is a huge thing as it can be used to dictate what is and isn't popular. Look at any market and there's marketing - look at fashion where the whims and choices of the market leaders dictates what is the "in" thing. Of course marketing has to follow as much as it has to lead and there are many examples of heavy marketing which utterly failed - 3D TVs and 3D films being a prime example where even with heavy marketing they didn't take off in a huge way.

Books at the same but I think also lacking the finances behind most publishing houses to really push books heavily as some other markets. At least until the book is a mainstream best seller - even then I've never seen a TV advert for a book (that wasn't wrapped up in something like morning TV shows). Used to be we got magazine adverts (They tend to hide now on niche channels) but books never so.



But I'd be willing to bet if you took a decently written book with standard modern english and plastered its details on every bus in London; in every newspaper and on the TV it would sell - Indeed get a popstar or similar to endorse the book and it could be away! OR at least have the potential to do so.



And thus the research comes in - the finding of that point where the content and style of the work is suitable enough for a company to invest heavily into its marketing to push it sales. Get it right and you get a Harry Potter - get it wrong and you get a 3D TV

 

[Phyrebrat]

Maybe analysing is productive in terms of user-friendliness of a novel, but really, I have to wonder if we've leaped ahead to April 1st.

I've never understood the almost asinine adherence to 'rules' in creative endeavours. The rules are there to support the creative artistry and innovation, not the other way round. It's like art houses using dot-to-dot or colour by numbers to qualify an oil painting.

What excites us about a novel? Does one read and think 'how well constructed this is!' Or does your head turn at beautiful prose?

I think it may be used to whittle out poorly edited submissions but then what would James Joyce have thought about that?!

We constantly read from agents and scouts that they look for something that excites them; something original. Structure doesn't bring that alone.

This just makes me think of terms like Facebook algorithms and Google Analytics and that hurts my soul!

pH

 

[HareBrain]

Yes, you can't algorithmicate for an emotional reaction. Good structure can certainly help build an emotional reaction -- through mounting tension, say -- but if a reader isn't already grabbed by characters or prose or setting or an idea that particularly resonates with them, it won't matter.

 

[Overread]

There is good structure and form to language; even an emotional passage can be pulled apart to understand the core flow and why it works both in isolation and within the context of the work as a whole. And of course there are multiple ways to achieve this; that doesn't mean you can't understand why it works, just that there are no singular best answers.

We know there are rules to language and writing, we use them every day. They are not shackles but tools that release creativity and allow writers, artists, musicians to communicate with their audience (or at least the greater part of their audience). Of course purely technical writing tends to be very flat, legal documents being a prime example of technical writing without emotion. But that doesn't mean there isn't a method to emotion; that there isn't structure nor form even if many of us use such methods purely instinctively rather than formally learned.

 

[Gonk the Insane]

publishing houses are trying to collect data to provide a basis for determining if a book will sell well, or not

On the surface, companies like Jellybooks mentioned in the article sound like they could be helpful, at least in as far as judging any weak points in a book, e.g. a chapter is so bad that readers stop reading. But... how is that really any different from the process of the various folks at agencies and publishing houses reading through manuscripts and finding those flaws and inconsistencies like characters acting out of character? Or beta readers, who would (hopefully!) also pick up on such things? I think it could be useful to self-pubbers/small houses to gauge readership, but unless you're looking at large numbers of readers, results would be of dubious value (plus there's the risk of writers re-writing sections of their novel to please readers rather than writing the book they want to write - that thought terrifies me!). And, even then, there would be those books that didn't get a good reaction but still proved remarkably popular once released. Maybe looking at reading data could help with the decision making process, but I'm not convinced it's any substitute for the good old-fashioned editor's gut (even though, yes, publishing houses get it wrong plenty of times).

I get the impression that a lot of editors are focused on the quality of the prose.

Uh, that doesn't sound like such a bad idea to me (practically sensible by publishing industry standards!). Structure is, of course, important but ultimately the writing has to carry the story. Great prose can bring to life a familiar plot (I think the ancient greeks thought that all stories were a variation on 7 core stories?), build tension, or make the mundane seem magical.

I'd probably go further, actually: if you can write good, solid prose then the chances are you've already acquired an understanding of structure and the elements that are required for a novel. And even if not, we read and watch so many stories that the elements of a comedy/tragedy/alien-love-affair are percolating below the surface of our minds. Conflict, I guess, would be the exception: so many of us starting out (guilty!) make our heroes too perfect/infallible and avoid plunging them into conflict with real, unguessable outcomes.

Certainly when it comes to open door submissions, if someone's looking at the opening chapters and the synopsis, I'd rather be judged on the prose than the synopsis. Getting someone interested in my characters and - by extension - their story is basically my job (well, not my day job, but you know what I mean).

you can't algorithmicate for an emotional reaction

Algorithmicate is my new favourite word!

 

[The Big Peat]

I think I don't agree on the possibilities of algorithmicating emotional reactions.

I think you'd need very good algorithms, a lot of data and a strong understanding of when the data comes into play and when its useless, but I think you could start getting something sort of useful. Not amazing, but useful.

But could you use ebook data to work out what sort of subplots appeal most to trad fantasy readers? What sort of MCs appeal? I think so. You'd need a huge amount of data to start getting something reliable but I think you could start being able to predict how readers react to certain situations with greater accuracy. Maybe you'd find out that moderate above the neck romantic subplots are preferred in bildungsromans and military fantasy, but people react badly to that in urban fantasy and political intrigue fantasy. Or that elves and grimdark result in unfinished books. Or something completely unexpected

Would that allow the publisher who knew that to have an advantage in spotting books that will sell? I think so. Yes, the book still has to be well written with engaging characters, but having a deeper understanding of what combination of events, people and so on resonates most should be an advantage in sifting out the good from the very good.

It could also result in the market getting locked even more heavily into "Do X, Y and Z or you're dead" but if I was a publisher who had the resources, I'd totally be looking into this sort of thing.


In any case - if I was looking for a bestseller today - I think what I'd be looking for most of all would be one big compelling hook that will make people go "Wow". You're looking for something that will make people go out and tell their friends about it. If its got that - be it a character like Glotka, or a place like Hogwarts, or whatever - the rest of it can frankly go hang. As long as its good enough no one's going "wtf", the "Wow" will carry it on its back. Without that "Wow", then everything else about a book can be very good, but people aren't going to fall in love with it in sufficient numbers.

I would be then looking for something that's clear and easy to understand. Clearly written, clearly structured, non-confusing characters. Nothing that distracts the reader from the "Wow".


This doesn't coincide well with the the value I potentially see in data I'll admit. Data won't teach you what "Wow" is. But that's okay. Data can help you sort through the things that aren't "Wow". If you look at sport, nobody needs data to tell you who's dominating the sport - it comes into working out the best of the rest. Data for publishers isn't about spotting the next Harry Potter per se, its about working out which Harry Potter clones to back.

 

[Andrew Lambert]

Wow. Really enjoyed this thread. Some great views about a world, up until 5 days ago, I had no knowledge of.

 

[S Blake-Smy]

The thing that scares me about these type of pushes, is that industries then start to ONLY produce what the algorithm tells them too, and we end up with formulaic genres that basically just repeat themselves and only the odd anomalous individuality starts to seep through.

Its like with pop music, you wouldn't necessarily get a rolling stones or David Bowie these days because it's more of a risk, and the algorithm says... no. (I'm being slightly simplistic in my terms but I hope you get what I mean)

 

[Jo Zebedee]

The thing that scares me about these type of pushes, is that industries then start to ONLY produce what the algorithm tells them too, and we end up with formulaic genres that basically just repeat themselves and only the odd anomalous individuality starts to seep through.

Its like with pop music, you wouldn't necessarily get a rolling stones or David Bowie these days because it's more of a risk, and the algorithm says... no. (I'm being slightly simplistic in my terms but I hope you get what I mean)

It is very much the case. The book of mine Brian mentioned above (Inish Carraig) fell down on demographs. (And, prehaps, an Alien invasion of Belfast Isn't mainstream....) But readers like it - so how do demographs either encourage something different into the market or take a risk when margins are so tight

 

[S Blake-Smy]

It is very much the case. The book of mine Brian mentioned above (Inish Carraig) fell down on demographs. (And, prehaps, an Alien invasion of Belfast Isn't mainstream....) But readers like it - so how do demographs either encourage something different into the market or take a risk when margins are so tight

I agree. It's a 'Culture of currency' rather than a 'Currency of culture' -- sounded good in my head. Anyway, I suppose it is inevitable, but I fear the stagnation that this scheme could encourage. All the money would go to those that just tick all the boxes rather than those that have something to say.

 

[The Big Peat]

Mm. When I make the case for data analytics from a publisher's point of view, I make it solely from their point of view. I do not think it will do literature as a whole much good.

I have books full of great flavour combinations. I use them frequently - but life would be sad if no one tried to add to them.

 

[Andrew Lambert]

It is very much the case. The book of mine Brian mentioned above (Inish Carraig) fell down on demographs. (And, prehaps, an Alien invasion of Belfast Isn't mainstream....) But readers like it - so how do demographs either encourage something different into the market or take a risk when margins are so tight

My Sci-Fi fantasy is based in Southern Ireland - Let's change their demographics perception!

 

[Jo Zebedee]

My Sci-Fi fantasy is based in Southern Ireland - Let's change their demographics perception!

Ireland as a fantasy setting is well established, never fear.

 

[Montero]

Wandering back to Harry Potter - JK Rowling did a lot of marketing. She went round from school to school giving presentations and getting kids excited about it and it spread from there. Those kids told other kids. So as others have mentioned - marketing does have a big impact. As does something becoming popular. Stephen Hawking's "A Brief History of Time" turned into a popular coffee table book, being bought by a lot of people who would have struggled to read it.
As for algorithms - suggests to me it could be a useful tool to apply for analysis if you are sitting there, staring at your manuscript, going "there is something there that isn't working". Not necessarily to then adjust your book to the algorithm, but it could highlight to you things you couldn't see because you were too close. But algorithms in general - wasn't there a stock market bond that was entirely based on an algorithm - that in turn had inbuilt assumptions about market behaviour. Did very well for a few years, then the market went outside the bounds of the algorithm and it all went to heck very, very, fast.

 

[MWagner]

However, I get the impression that a lot of editors are focused on the quality of the prose.

I think they use the quality of the prose as a measuring stick. A submission goes in the bin if the prose quality is below a threshold, because revising a book to bring it up to an acceptable level is too much work. Structure is important. But you don't know if a book is well structured until you've read the entire work. And from what I understand, publishers don't have the time to do that for more than a small fraction of the submissions that come their way.

My impression is that most of the policies in publishing are necessitated by extremely limited resources, and are intended to winnow down a cascade of submissions to a small and manageable number that get serious attention.

 

[Stephen Palmer]

Editors want authors they believe they can work with. That means an author with no ego issues, who does the work and can write English language without too much problem. They also want the buzz of reading something with that indefinable magic, and they're mostly worried about accidentally missing the Next Big Thing.

 

[Overread]

I think they also want authors with a certain writing speed as well. Sure if you get big like GRRM you can afford delays because a poor product won't sell and because his back log of previous works will still sell in high volumes. However authors who are very very slow to write or who can't put out a reliable stream of product might well find it hard to get contracts or to keep them. Publishers don't just want one book they want more - sometimes I think to the detriment of the world setting. Look at Pern for something that clearly got stretched out, partly because it sold, or at authors like Raymond E.Fiest who clearly lost some of that spark in some of his latter works.

Although it should also be said that I'm unsure how much of this was market pressure; publisher pressure and the authors own choice. Market and publisher pressure could be subtle - your new line of books not getting the same advertising and marketing chances (look at Rowling - her non-potter book hardly got the masses of publicity that any new Harry book would get). Thus an author could be pushed into working on their initial success.

Indeed if we follow that line of thought its actually quite amazing that for the skill many authors have most are known only for their skill in writing one, maybe two signature series. Some are more crafty than others with that - Terry Pratchett adapted discworld to let him write quite a few different characters, voices and themes into his work whilst maintaining his general style and world setting.


If often seems that to be a more modest author who sells but isn't leading is actually of benefit if you don't want to be stuck to a single series/theme as it seems that the pressure just isn't there in the same way to stick to a single series.