The real action is with the store inventory database, both for compiling best seller lists and tracking consumer choices. They totally outclass the credit cards issuers when it comes to monitoring what you buy. Cardholders do break purchases down into categories, like restaurant, retail, etc.

I’m pretty confident of my comments, but it wouldn’t hurt to double check them if you’re really interested.

Now that I think about it a little more…is it a given that credit card companies *really* know such ISBN specific data?

I mean, if I go to a Barnes & Noble on Tuesday and buy Book X for Amount Y and pay with my credit card, does the credit card company really know that I bought Book X? Or do they only know that I spent Amount Y at such-and-such bookstore on such-and-such time?

Obviously, the bookstore knows what exactly I bought and when and for how much and that I paid via a credit card. But it seems to me very possible that in the normal transaction process, the credit card company might only knows how much I spent and where, but not necessarily what exactly I bought. (That is, the credit card company might be asked for an authorization for such-and-such amount, but they might not know the exact goods and services that make up that amount.)

At this point, this is very much a digression, but now that you’ve got me thinking, I do wonder about what exact information credit card companies routinely capture.

“For my personal use, unless they are getting their numbers from all the big companies (which they don’t) and a good sample of small retail shops, I find it hard to trust their numbers.”

I think perhaps the best take-away from all this discussion is a reminder that EVERY best-seller list has its own nuances, flaws, biases, etc. and that they’re best viewed in that light. In the specific case of this New York Times list that’s started all this conversation, the New York Times of course has had a book bestseller list for forever. If one accepts that that list is basically OK (which maybe one doesn’t; every list has its problems,) then that implies that they’ve got acceptable coverage of “real” bookstores.

If we assume that they’re getting good enough data from “real” bookstores, the real question with their new graphic novel list boils down to how they’re covering the comic book specialty stores. That further breaks down into the sub-questions of (1) whether or not they’re getting sufficient amounts of good data from comic stores to make for relevant analysis (we know from retailer comments on other blogs that some comics stores are indeed capturing and reporting this data, but are enough of them?) and (2) whether or not the math they’re using to integrate the “real” bookstore data and the comic store data yields an accurate picture.

I kinda doubt they’d ever make public the answers to these sorts of questions, but I like to think that they’re reading all these internet comments and are taking them into consideration as they refine their methodology in future.

While I am open to the idea that it is technically possible that comic stores and book chains could share their sales data, it seems unlikely that (in general) they would be willing to. I’ve known a few comic shop owners and keeping things daily in the black would be far more important to them than worrying about contributing their sales information toward best-seller lists. Perhaps the data collectors somehow make it worth their while.

For my personal use, unless they are getting their numbers from all the big companies (which they don’t) and a good sample of small retail shops, I find it hard to trust their numbers.