Tilting at Windmills #119: The Bookstore Equation: 2003


by Brian Hibbs

[Originally publlished #119 – February 2004 – “The Bookstore Equation: 2003”]

I suspect the “hot topics” in the current internet comics-industry commentary can be boiled down to one of two largely intertwined positions: 1. Bookstores = Salvation and/or 2. Manga = Salvation.

The problem is I think much of these positions are effectively based upon a game of Telephone.

We started discussing this last month – we’re just not seeing enough real and meaningful data to make many effective conclusions, and what data we do have is often misinterpreted.

So, here’s an attempt to interject a few more facts into the mix. Not all of them, mind you, and you may find my conclusions to be just as dodgy – but I can give you one more piece of the puzzle than you had yesterday, at least.


If you go over here you can find a file with the full BookScan listing for the last week of 2003. If you’re a numbers-wonk, like I am, I encourage you to grab a copy of this yourself. Quite technically I suspect this is “proprietary information”, but given that I’ve received no less than three copies of this data from various sources, I’m going to file this one under “journalistic privilege”.

Besides, information wants to be free.

So, let’s handle a few background-style questions about what this is and isn’t before we get to any analysis. The impatient among you will probably start scrolling down (this is the internet, after all), but I think we should all at least try to be on the same page.

First off: What is BookScan? Well, they’re a division of Nielsen, the people who bring you television ratings. What they do in the book market is collect and present POS (point-of-sale – those bar-code scanner thingies) on every single ISBN number sold through the retailers they use.

In an e-mail conversation with one of the VPs of the service, she indicated that they track each and every ISBN, “except for some non-book product that carries an ISBN such as diaries, book marks, book covers, etc.” Thus, if a graphic novel or trade paperback is sold through a BookScan retailer, that sale should be tracked.

But who are the retailers who report to BookScan? According to the list that I have, there are over 7400 potential BookScan venues. This list includes almost 300 independent bookstores, as well as chain retailers, B. Dalton / Barnes and Noble, Borders / Waldenbooks, Tower Music and Books, Musicland, Deseret Book Company (Mormon bookstores), Follett Stores (University bookstores), Hastings, Costco, K-Mart, and Target. BookScan also tracks online sales from Amazon.com, B&N.com, Borders.com, Buy.com, Fatbrain.com, and Powells.com.

That’s still a fair number of places that sell our product that aren’t represented – beyond traditional book retailers who don’t report to BookScan (Say, Baker & Taylor, or the rest of the indie bookstores), and mass market retailers like Wal-Mart, this is probably missing a big chunk of library sales, university sales, airport sales, etc. This Publisher’s Weekly article (you’ll have to subscribe to read it, sorry) says the following:

BookScan generally claims to represent between 70% and 75% of sales in the industry (Wal-Mart and some of the supermarket chains are among those who decline to report.) But a comparison with in-print figures supplied by publishers reveals that the numbers are more likely to represent about 65%, even after deducting for unsold books and returns.

For BookScan’s top ten nonfiction titles published last year – a list that include mass-market favorites like Phil McGraw’s diet books as well as indie hits like Benjamin Franklin: An American Life – no title had BookScan sales comprise more than 75% of total sales. For some of the books that had strong special-sales, they ran as low as 25%.

To, sadly, add a little more noise to the signal, you can’t really find an internal-to-comics common response from publishers as to how well BookScan tracks. One publisher of a large comics company indicated he thought BookScan numbers ran higher than 75% for us — traditional bookstore chains being historically stronger supporters of the TP than the more general “mass” stores.

In any event, you can add an unknowable, but almost certainly significant, number of copies to what’s listed in this chart. Of course, you can say that about the estimates of the Diamond charts, too.

Right, so what the hell is this file, anyway?

This is the top 750 sellers for the week of 12/31/2003. This is not an “end of the year Bestsellers” list, except accidentally. In order to get on this list, a title needed to sell no less than 68 copies among the approx. 7400 BookScan reporting stores, during the last week of the year. It is, as presented, currently ranked by copies sold that week, but I think one exercise anyone should perform is sort it by Year-to-Date (YTD) as well as by the column rather disingenuously labeled “publisher”. (Really, that should be “distributor”, for those keeping score at home)

When comparing the BookScan numbers to the DM numbers we have, I think it’s clear that the DM usually sells as many copies of “traditional” comics material, if not far more. This is despite having fewer venues – Diamond has said that something on the order of 3300 accounts turned in an order form in an average month in 2003, and, interestingly enough, only about 1800 ordered backlist material (i.e., something with a STAR code) during the (arbitrarily chosen) month of September 2003. While trying to figure out how many BookScan reporters are selling comics is almost certainly an exercise in futility, I think it’s probably safe to say that, in terms of raw number of stores, the DM has at best no more than half the number of venues selling TP- and GN-format comics compared to the book market.

The Direct Market’s buying power is clearly the engine driving the bus – those firm, non-returnable, sales are absolutely essential to the production of comics in America, and if the DM were to “go away” tomorrow, I think that it is clear that the bookstores would not be able to pick up the slack.

One thing when looking at the Diamond numbers – there are a tremendous number of sales that simply don’t show up at all. For a really good example of this phenomenon, look at the Diamond Year-End report for TP/GNs: Watchmen comes in at #23 for the year over all. The item at #22 on the year-end is Trigun vol 1. ICv2 estimates sales as 6889 in October, 2981 in November, and 1691 in December – that’s 11,561 copies cumulative. This means that it’s safe to say that Watchmen almost certainly sold at least 11,000 copies in the Direct Market despite never once showing up on any monthly Top 50 list!

Think about that one for a moment, eh?

So to bottom line this discussion: most of the data we have is flawed or faulty, and generally speaking, we’re comparing one very flawed reporting method to a second very flawed reporting method. While I believe that some general shapes can be discerned, I am very mindful of the perversity that I’m comparing apples to sardines, and I don’t even know what an apple looks like, really. However, we do the best that we can.

So how, specifically, does the DM compare to the BookScan report? Well, here are some of the things I see….

The Direct Market sells many things equal or better than the bookstores – even things they’re “supposed” to sell best. Don’t think so? Well, how about that Sandman: Endless Nights? 28,710 in book stores, 33,592 (26,230 in September, 7362 in October) in the Direct Market. Less stores, less promotional budget, less just about everything and we beat them by nearly 5000 copies! New York Times Bestselling author and all that, and we specialists are shifting more copies.

Buffy creator Joss Whedon? His Fray says 687 in the book stores and 4210 in the DM (And, sorry, I’m not linking to every ICv2 figure from this point on – far too much work!)

Matrix Comics? Bookscan reports 8171, while ICv2 has 8285. That’s astonishing parity for the DM, especially considering the book was ordered not through Previews, but as a Previews Update, which usually generates much lower orders than a listing in the full catalog.

Books about our icons sell better than the icons themselves in the bookstores. DK Publishing’s Superman: The Ultimate Guide shifted 10,852 copies, while The Return of Superman TP was the comics best-seller with a measly 3103. Those are both BookScan numbers. DK’s Spider-Man book did 15,342, while the best in Spidey comics is 11,460 for Ultimate Spider-Man v1. DK’s X-Men book hit 41,040, blowing even Origin’s 25,601 away. And so on.

Humor kicks the most ass in the bookstores. The number one book for the year? Nearly 77k copies of Get Fuzzy. Huh, it was also #2. FoxTrot, Boondocks, Zits, Far Side, and Dilbert all sold better than almost anything else. Even Calvin & Hobbes, which ceased publication as a new strip eight years ago, devastates most of the competition.

What I love about this is how the pundits say we should be more like manga, while no one ever suggests we should move more towards humor comics format. Why not? They rip up everything else in sales. If only we were to switch everything over to a four panel gag format, we’d solve all of our problems, right?

Also of note is the strong success of Bongo’s collections of The Simpsons. These books are completely invisible in the average Direct Market store, yet they sell tens of thousands in the book market.

The existence of film versions don’t appear to have a great impact upon comics sales, in and of themselves. You’ll note few appearances of Hulk or Daredevil, and what they do have is very weak by Direct Market standards. “Oh, but those movies stank!” you say. Well, so did the League of Extraordinary Gentlemen, yet the comic is in the top 10. Hulk’s box office was 2:1 of LXG.

“Art” Comics don’t really sell in bookstores, either. At least the venues BookScan tracks. Extraordinarily few “art” titles (call it the D&Q/FBI/Top Shelf axis) show up at all, and what is there is generally way down the charts. So much for any real hope that the “mass” market would embrace the bleeding edge of what we produce. In fact, unless I’m counting wrong, I only see 19 titles that aren’t either manga, humor, or from/about one of the “big four” publishers. And not one of those nineteen cracks 10k copies.

However, one smaller publisher said to me that “on ‘A’ titles the bookstore market can outsell the DM 8:1, but on everything else, the DM is 8:1 over the bookstores”, and that refrain, or something very similar to it, has been consistently repeated to me from a score of publishers.

Remember: cracking into the book stores isn’t easy at all, in the first place. And even if you do get your foot in the door, you have to deal with the possibility of returns, which has been at least one factor in the downfall of many small publishers over the years.

Bottom line: the book market is still only a fraction of the DM, and much of the material that we most want to see succeed really only does so in the DM.

Lest you misunderstand my intentions, I really want a strong bookstore market. We had a decade or two without any appreciable “feeder” market for the specialists, and that’s a terrible thing. If strong sales are engendered in the bookstores, that can’t help but aid the Direct Market.

A generalist market is only going to stock and sell a general selection of material. While Watchmen and League of Extraordinary Gentlemen both did well in the book stores, we’re not seeing Alan Moore’s other work showing up on the BookScan charts. Speaking anecdotally, I know I’m doing very well with the rest of the ABC line (Promethea and Top Ten especially); with Checker books reprints of his Awesome work; with Swamp Thing and V For Vendetta – and I assume that other DM specialists are as well. The book market, apparently, is not.

The book market may in fact be growing (hard to tell from one report), but it’s not salvation, and it’s not the Promised Land, and while they may someday sell more comics than we do (they have more venues, after all), that is going to be a long time coming.

* * * * *

So, what about the manga side of the question? Clearly it is the category killer on BookScan: 55% of the top 750 are manga!

Some of them have massively impressive YTD sales, too – 48k on Yu-Gi-Oh, 39k on Chobits; however, over three-quarters of the titles listed have sales of under 10k, and nearly half are under 5k.

The book stores approach manga very differently than the DM. As we discussed last month in the DM when we order comic books we’re betting non-refundable dollars that our orders are right. If DM stores guess wrong too often, they go out of business. This means that retailer confidence is the key for the DM

Now, most DM retailers simply don’t understand manga. And it’s not because of some sort of Simpson’s “Comic Book Guy” myopia – “we hate this because it is new” — it’s because it’s literally foreign to what we know and sell.

Look, I don’t know what a “Rurouni Kenshin” is – I don’t even know how to pronounce it. I couldn’t tell you if it was any good, or who it is meant for, or anything about it, because it is simply outside my circle of knowledge of western comics.

It wouldn’t be so bad if it was just one book – we can learn. But it’s not one book, nor a dozen, nor a score, nor even a gross. It’s hundreds upon hundreds of titles.

Let me give you a one-month snap-shot of the problem: on the January ’04 Previews order form; there are 18 new TPs from DC. There are 16 from Marvel.

There are 29 from Tokyopop, and 27 from Viz.

Each of those companies is coming dangerously close to exceeding the output of Marvel and DC combined. Think about that for a second! When you read the feel-good notes that manga has increased in sales by 75-100% this year, it’s worth noting that production has certainly increased by the same amount, or more.

Now, at $5.49 (my cost) a copy for a manga trade that I really don’t know anything about, and have not been able to find a definable customer for, that’s an awfully expensive education if I get it wrong.

So I order one copy each here and there, trying to reorder what sells, trying to liquidate what doesn’t. And, at the end of the day, my manga section ends up being a wash for me – I liquidate more books from that category than any other I carry, and I’m still looking for that elusive manga title that I can turn more than, say, eight times a year. I turn something like Watchmen thirty times a year, and I consider that a solidly “midlist” book.

This year there are probably going to be something over 1000 new manga releases. I honestly don’t see how either market is going to absorb all of those, so what I think we’ll see a continuation of the trend that a few books score stellar sales, while the mid- and lower-list titles fall off rapidly.

The question is will the coming glut self-correct itself before it begins to take businesses out? Whether that will be over-producing and over-extended publishers, or overbuying retailers, a correction will come. It always does. The only ones really not at any significant risk are the bookstores, who have returnablity.

We can’t tell from a one-time chart if manga is a “fad” or not – fads choke on over-production, while new categories correct then move forward – and we probably won’t even be able to tell by 2004’s numbers. But by 2005, we’ll see if the category has real “legs”, that is, if we settle into a large group of books that are selling at least 10-20k per year, every year, until the end of time (Like Watchmen, or Sandman or even Ultimate Spider-Man) Based upon the limited snapshot here I have my doubts, but I’d be happy to be proven wrong.

Theoretically, manga is being sold to a demographic largely not being serviced by DM retailers – teens and tweens, especially young girls. (The reason that these demographics aren’t being well-served is because comic book stores are largely destination stores. You have to seek them out to find them.)

However, this demographic is another reason why I hedge my bets in declaring any victory for manga – teens and tweens are notoriously fickle and disloyal, and very seldom do you get something with long-term bounce from that bracket.

One thing I do know is that the “manga-ization” of western comics is probably a lousy idea. We only have a few examples we can clearly show in 2003, but each of those show that the DM did better than the bookstores. Despite the bookstore being the primary market for manga.

The most visible one is probably Death: At Death’s Door. BookScan shows 9846 copies sold, while ICv2 posits a total of 19,582 copies in the DM – nearly 2:1.

We can also point to the reformatted Elfquest (967 versus 3927 for v1 and 452 versus 2923 for v2 – in both cases in favor of the DM) or to Powerpuff Girls. This one I find almost embarrassing given the Bongo Simpsons domination on BookScan. “Titans of Townville” only tracked 655 copies on BookScan, YTD, while ICv2 claims 2525 in the DM.

There’s also Batman: Child of Dreams, though that wasn’t published in manga format, so maybe it’s not a great example. 3746 in the DM, 503 by BookScan.

Either way, I wouldn’t rush to making our product more “eastern”. That’d be a knee-jerk reaction to a market trend that’s just starting to play out, and which can honestly go either way. It will be easy to “glut” the consumer away, like we’ve done so many other times in comics before – and, in fact, many signs are pointing that way right now. I certainly don’t want to see manga collapse, because if there are legitimate long-term customers there that can only be good for comics. However, the evidence seems to be that they want real and proper manga, not western comics “manga-ized”.

Finally, I think there’s a trend in manga that hasn’t really been discussed publicly yet. My problem is: I can’t point to a number on either chart that illustrates this directly. But my anecdotal experience is that many manga TPs aren’t actually selling as perennial backlist items, but rather have a sales pattern much more akin to periodicals. That is to say that they sell strong during their release and while new volumes are coming out, but die a dog’s death on the rack once the whole series is complete.

Like I said, I can’t prove this, it’s purely anecdotal, and we’d need several years of comparative data to establish it with much certainty, but, if that’s actually true, then that goes a long way to “explaining” the sales of Love Hina or Chobits – the phenomenon is that we’ve effectively swapped one kind of periodical for another in the “mass” market.

See, the key to a graphic novel driven market lies in the “turn” – how many copies of a book can you depend upon selling month in and month out. This gives you the economic base to sustain your business upon, and to grow from. If I can depend on selling 2-3 copies of Watchmen each and every month that’s, say, $600 a year that I can count on to pay my bills. The path we want to follow is one where we add a wider and more varied backlist that consistently turns. Those 30 copies of Watchmen I sell every year don’t mean much in and of themselves – but when you add it to the year-in, year-out sales of Sandman, and Cerebus, and Bone, and Transmetropolitan and Ultimate Spider-Man and Sin City and Love & Rockets and so on, suddenly you have a significant economic engine. (An engine that, might I remind you, is almost completely invisible in current DM reporting methods)

The economic process behind periodical releases is vastly different. In that model, initial release is where almost all of the sales are. Then things begin to drop off sharply, so you’re trading on your ability to gauge sales in advance.

We may be comparing the wrong things when we look at bookstore manga sales. It might be that we should be comparing most new manga sales to the monthly comics charts – still amazing figures even by that comparison, but not quite as overwhelmingly out-of-balance.

(And, frankly, I think this is the year to be very careful of manga, and to weed ruthlessly, rather than expand recklessly. Advice to the contrary makes me think its 1986 and the suggestion is that one should move big into new b&w #1’s. Or that it is 1993, and you should be stocking your store with foil covers.)

As I say, this is the trend I think I am observing at Comix Experience – I’m not sure I’ve sold more than 2 copies of Love Hina or Chobits since they finished, just like I don’t think I’ll be moving many individual copies of 1602 once that mini-series concludes. I can’t give you more than anecdotal evidence, there’s nothing in the charts I can specifically point to and say “See, there!”, but my gut tells me this is the way the cards are playing out.

And this will only accelerate as the amount of manga produced doubles again this year.


Brian Hibbs has owned and operated Comix Experience in San Francisco since 1989. Feel free to e-mail him with any comments. You can also purchase a collection of the first one hundred Tilting at Windmills (originally serialized in Comics Retailer magazine) from IDW Publishing. Oh, and by the way, can you name what Tokyopop’s best selling title was last year? Nope, you’re wrong – it was the fumetti-style Lizzie McGuire v1. Not manga at all.