Log in

No account? Create an account

Previous Entry | Next Entry

Bestsellers again

So the pernicious idea that all publishers are now pursuing a model in which "bestseller or bust" is the anthem (I want goose-stepping, too) must come from somewhere.

(Oh, very quickly, the idea that if your book isn't a bestseller, it gets whisked away to the backroom at the end of the month to suffer a gloomy fate is kind of funny, because generally speaking even if it is a bestseller, it gets whisked away to the backroom at the end of the month to suffer etc.

There is always a bestselling paperback or dozen, and there are always placement dollars for those books that necessitate the previous month's bestseller's removal from prime retail real estate. I covered this a couple of years ago, in a) returns and b) promotion – but it bears some (small) repetition here: To keep multiple pockets of a novel on display you need to have a minimum number of books in stock. Those books are essentially filler. You don't really expect them to sell; you expect them to be there so that the book has prominent display (which was paid for by the publishers)(unless head office didn't order enough for that display, which does happen, and boy would I be furious if I had paid money for placement).

Given the space for backlist, you are unlikely to keep all of those books in your store when you can always reorder should they sell out, so you strip them and return the covers. The timing for hardcovers used to be longer, but there is so much published in hardcover these days that hardcovers do get pushed along, just like everything else.

That was this post's digression.)

I think that the notion of "bestseller or bust" comes partly from writers and agents and the stories that arise from midlist sales.

By this point you've all heard of authors whose books were published to low numbers, and whose subsequent books, thanks to the vagaries of computer ordering, were published at even lower numbers in a kind of downward spiral, the result of which was that said authors were frequently dropped by their publishers. These stories are by and large true. On an individual level, your book needs to develop legs quickly – by legs, I mean needs to sell out and be reordered fast. At one point twenty years ago, no one expected this kind of instant momentum, but twenty years ago is just that.

Ergo, if you personally are not a bestseller, your book(s) die and you fade from view. Building this up into a situation in which "bestseller or bust" seems the rule is not that hard to do, and I believe that that is what's being done in this case – but from a publishing perspective, it's not mathematically accurate.

Because: While the individual midlist authors may come and go because of the way orders & numbers work overall, the fact is that the sales numbers for the midlist in general are not actually dependent on specific authors. When one midlist author has been let go, another has been found; the publisher soldiers on, even if individual authors pause by the road-side to take stock. The publishers can make money on a reasonable midlist populated by, well, midlist writers; they don't need to spend 9 billion dollars chasing the next Harry Potter. And for the most part, they don't.

And even in the event that an author does attain Bestseller status, you needn't look very far to find one who was bought as a midlist author – Rowling, for instance, with her first UK publisher.

Okay, I feel a pressing need to digress again. Because I think a point should be made about sales numbers. It's not necessarily the numeric value of a single title's sale that causes excitement among sales reps.

Take two midlist first novels. One has a laydown of 25K copies in mass market. The book eventually has a decent sellthrough, but in the end, the second book will not necessarily be ordered in the first book's initial numbers, so a downward or static trend is inevitable. But the other book… well, that book had an initial laydown of only 20K copies. And people are hyped about it. Why?

Ms. Big Chain Buyer, orders 20,000 copies of the second First Novel. Of those 20,000 copies, an absolute minimum – regardless – of 3000 books will be returned as overstock, and this would in general be considered a fabulous sell through. Why? Because what will generally happen in this case is that most of your outlets will have sold out of the book, but owing to geography or shelving, etc., some won't, and will duly return them.

If both authors are in this position, so far so good. However…

IF the outlets that do sell out manage to sell out quickly enough and IF people come in and special order the First Novel, etc., the book will be reordered (I leave off the cases in which the book is underprinted and is in a 2nd or 3rd printing before the book even hits the shelf, generalizing for the sake of simplicity). Most big houses can resupply the book in 2 or 3 days if they have any in stock. So while my initial laydown was 20,000, if the book actually sold really well and was reordered frequently, I might end up with orders for 25,000, and returns of 5,000 in the first reporting period.

That 25% figure that gets tacked on in reorders will influence the number of books BCB will order by that author when the Second Novel is offered. And everyone in-house knows this. The next laydown for the author's second novel is likely to be higher than the first (say, 30K conservatively), with backlist resoliciting of the first book.

This is, of course, momentum. This is what people, in the end, really want to see. You don't need to be selling 300k copies to generate that level of excitement.


( 8 comments — Leave a comment )
Oct. 20th, 2006 02:39 am (UTC)
You're the first person I've seen talk about this within (I'm generalizing Fantasy) Genre; most authors I've talked to write Canadian Literature and if they sell a quarter of the runs you stated it's apparently a good sell, but I wouldn't know from personal experience or anything to that effect.

My question is for generalized time of shelf-life you've found for the average midlist novel. I know it's hard not to generalize and say for sure, but this has piqued my interest.
Oct. 20th, 2006 03:19 am (UTC)
7500 copies shipped of a Canadian novel used to be considered almost bestseller status many years ago. And yes, my gross generalizations really refer to books published by a US (genre) publisher for North American distribution.

I do not have hard numbers on the average midlist novel's life -- but I think it's actually longer than it used to be in the 80's. Books don't seem to go out of print as quickly as they did before, and whether this is due to more organized warehousing, or the Big Book Stores that require a huge amount of backlist (as opposed to frontlist) -- which would make sense to me -- I'm not sure. I think it's a good thing, obviously, because it happens far less often that book one of a series is out of print while book 2 and 3 are still in the pipeline, which means that the sales of book 2 or 3 don't immediately wither and die.

And the numbers used there were entirely theoretical, and a midlist mass market can frequently see print runs of 9,000 copies in this day and age (in the US, for North American distribution). More titles, fewer books of each title -- same old story =/. I generally tend to use round numbers as examples because it's easier. That print run will, of course, be shipped out and returns will be factored in.

I know that my West novels have against all odds been kept in print (they're kind of... huge) since, ummm, 1998, I think (they're all upstairs, where people are sleeping, and I'm downstairs), in mass market. It is easier to keep trade paperbacks in print because the cost of doing a small run per unit produced vs. the retail price (and therefore publisher margin) is better than it is for a small run of mass markets (where the unit cost is measured in thousands).
Oct. 21st, 2006 04:20 am (UTC)
There's a lot more book metadata available, and available relatively easily unless it's about a highly popular book (in which case the noise level overwhelms all) than even a decade ago. The Internet's made a huge difference, the velocity of information flow, the availability of information, the distribution of information, and the amount of information, have all increased enormously. "Obscure" new books don't stay obscure if even one competent influential Internet blabberblogger find a copy and likes it enough to mention it glowingly to any of the audiences who trust blabberblogger's judgment/taste, or are willing to give books that blabberblogger suggests, a try.

As for the more organized warehousing, the reason why Walmart's the biggest retailer in the USA is its inventory control system. It's the metadata that makes the huge difference. The faster the merchandise turns around, the higher the revenue per square foot of store space. But Just In Time supply chain concepts, require accurate up-to-date inventory systems, and accurate (or at least reasonable projection) forecasting.

The regime where Made in China is eventually going to get clobbered, is when the currency rates change and the purely monetary "the workers there get $2 a day, workers in most of North America that's a few minute of time" advantage erodes away, and that slow boat across the Pacific and having to order months in advance, means that with the monetary advantage eroded, having to forecast demand months ahead of time, put in the orders, and then hope that the ships arrive, the port facilities are available (Walmart got temporarily clobbered nearly two years or two ago with merchandise stuck on ships that couldn't get into the overcrowded ship unloading facilities to offload Christmas season merchandise), the weather cooperates, and that what looked sensible to order six months in advance, turned out to have the predicted demand.

Having to make firm orders months ahead of time, doesn't make for "agile" business operations. One of the issues with publishing can be the lead time involved. I keep wondering when the contemporary fantasy vampires/shapeshifters/witches market saturates. IIRC Making Light had a thread many months ago about how fast a YA or children's series that had been having extremely high sales, suddenly quit selling, much to the detriment of the publisher's finances. Hmm, thinking about it now, I wonder how much that disrupted book retailers' operations and finances (though since it was only one series of books for retailers, it wouldn't have the same impact as it had on a publisher for whom the series was a big cash cow.)

Oct. 21st, 2006 04:44 am (UTC)
For frontlist titles -- the bestsellers in question -- Having to make firm orders months ahead of time is the rule rather than the exception. The stores, if they receive too little of a title, can reorder (sometimes; there have been times in Canada where no ordering at all was allowed by the individual stores. Hell, there was a period where no new books were being accepted either; I don't know if this has ever happened Stateside).

Given fiction and its lead-time, I'm not sure that publishing has ever been described as "agile" in its business operations. Lead-time can be cut to almost nothing for very timely non-fiction, but non-fiction is an entirely different species of publishing, and one about which I know little.

But in your model, the backlist is where good inventory management would show the best results -- and I believe that backlist is often decided by head offices (i.e. they'll tell you what books you're to keep on your shelves, and in what quantities, once they've seen how it sells otherwise).

As for metadata, I'm embarrassed to say I'm not quite sure what you mean. I'm not sure how to measure obscurity in a tangible way. The internet allows for much faster word-of-mouth, which is generally considered to be the second most important factor in a book's sale (the first, oddly enough, being previous enjoyment of a different work by the author of the book in question).

But in terms of quantifiable numbers, I still don't see much of that.

I was surprised that Bookscan, which is used in so very many outlets, reported about half of a particular books sales (I asked someone who works at a publisher who also has access to bookscan information) when compared to the royalty statement (and we may assume that on a royalty statement, a publisher is highly unlikely to inflate the numbers of books counted as "sold").
Oct. 21st, 2006 05:49 am (UTC)
"Metadata" is information about a topic, about where to find information, etc. A point to B which points to C where C is the information you're really looking for, A and B tell you where/how to find it. A and B are metadata.

(or, getting horrendously geeky googling...

[Error: Irreparable invalid markup ('<a [...] metadata">') in entry. Owner must fix manually. Raw contents below.]

"Metadata" is information about a topic, about where to find information, etc. A point to B which points to C where C is the information you're really looking for, A and B tell you where/how to find it. A and B are metadata.

(or, getting horrendously geeky googling...

<a href="http://walrus.wr.usgs.gov/infobank/programs/html/definition/fmeta.html>http://walrus.wr.usgs.gov/infobank/programs/html/definition/fmeta.html</a>

<b>USGS CMG "Formal Metadata" Definition</B>


<i>"Metadata" is information about data or other information.
"Formal Metadata" is metadata that follows an FGDC approved standard that provides a common set of terminology, definitions, and information about values to be provided. </i>

<i>Formal metadata are formally-structured documentation of digital data products. </i>

<i>They describe the "who, what, where, when, why, and how" of every aspect of the data.</i>

<i>Formal metadata: </i>

<i>help organize and maintain an organization's internal investment in data,
provide information to data catalogs and clearinghouses (e.g., USGS GD or National), and provide information to aid data transfers. </i>
<i>Metadata should be recorded during data acquisition, processing, and analysis -- when the information needed for metadata is known, not after the fact, when important information may be lost or forgotten. </i>

<i>Other metadata information also occurs in the CMG InfoBank for: navigation, gravity, magnetics, bathymetry, seismic, cameras, crew, geodetic postioning, lines, metering, ports, samples, and stations. </i>


Few things are truly pure supply and demand. Things like payment for product placement do skewing, for example, by preferentially locating merchandise where customers are most likely to notice, and in larger quantity than pure supply/demand models assume. Decisions by management on what gets to shelves and in what quantities limits the options available to buyers in stores who browse and buy what's in the store... and for someone to order something, they person needs to have an awareness that something to order, actually exists.

That's where the metadata comes in, someone hears about <i>House War</i> by Michelle West and goes looking around to see if the person can find a copy of this book they have heard about.

A whole bunch of enticements are around to get information out about books--sample chapters on-line are one of them, so that someone reading the sample chapters, doesn't have to buy the book to see what it's like. If the person reads the sample chapters and likes them, that's a fairly powerful promotional technique to get someone to chase the book. Also, recommendations in weblogs and otherwise on-line get people informed of books (including what's likely to not appeal to them--that makes for a more efficient market, because fewer books are likely to sell to people who are going to find that the book isn't what they expected and disappointed them. Instead, they can concentrate their attention for tracking down and buying books that seem more interesting/entertaining to them.

I don't understand why more bookstores don't provide customers the ability ti looking up books to get enough information to ask a store clerk if the store has Particular Book or first look on likely shelves--and that stores don't emphasize "if you don't see it here, you can still order it"

I think that best that can be said for certain book business inventory handling and tracking practicis is that tey're baroque. Books aren't laundry detergent and sales techniques for laundry detergent aren't optimal for promoting and selling books. February's book releases not interchangeable with September ones, isn't relevant to laundry detergent sales. If someone finds a detergent brand they like, they tend to stick with it and go looking for it. Books, however, again, are not detergent, and are supposed to be unique... and that makes finding one that came out months before, very difficult. The same thing isn't assured to be there in another six weeks, for books, versus laundry detergent.

(Deleted comment)
Oct. 20th, 2006 11:04 pm (UTC)
Hey :).

Is your WFC Dancecard full?
(Deleted comment)
( 8 comments — Leave a comment )