study on piracy and sales of ebooks

[piranha]

need to find out more; this is just a marker:

magellan media partners are doing an ongoing study of o'reilly titles and the effects of piracy on sales, and the correlations look very interesting so far.

for example, from moment of first seeding, pirated books shot up in sales to higher than initial release.

the study is for sale for a mere U$99 -- at those prices i think i might look for a pirated copy, *snrk*. but i can download a PPT presentation if i sign up with my email address. ok.

via cory doctorow on twitter, here's the "moneyshot" from a presentation yesterday: http://tinyurl.com/yzlnj8e

Comments

[maize]

I only looked at the one slide that you linked, so maybe I'm lacking content. I'm also feeling a little slow because the comments on that slide seem to imply that it has a clear and immediate meaning, whereas I'm not finding that to be the case without a lot more information (which I assume was given in the verbal presentation).

Am I assuming right in that the red line indicates "average sales for books for which we see no pirated copy during the course of our study," and the blue line indicates "average sales for books for which we found pirated copies during the course of our study"?

I'm not sure how the results, if that's correct, are all THAT interesting. It seems to show fairly planely that books that suffer piracy sell significantly worse than books that do not. Yes, there is a very short spike right after the seeded copy is first found, but it's incredibly brief and sales immediately drop back down to pre-spike numbers. Perhaps that does indicate some piracy-related sales, say, from people reviewing the book or recommending the book to friends, but the sales for the pirated books are so much worse anyway, that I can't see why that's all that relevant or interesting.

"Sales are much worse overall, but you'll see a momentary spike in your terrible sales at some point, after which they'll go back to being terrible again, and you'll suffer a net loss," doesn't seem a riveting argument to make in terms of piracy, given that I think that other than the almost meaningless momentary spike, that's the "party line" about piracy that people expect to hear.

[piranha]

i said that this was only a marker and that i needed to find out more for a reason.

i'm ignoring the data that's marked as "unreliably small sample set" for now (there are all sorts of reasons why some books might not sell well in the longer term; i'd really like to see a breakout along the lines cory is suggesting). it's the spike i find interesting, because that confirms my own anecdotal data. you may feel free to find this entirely uninteresting. :)

[crantz]

You know, it's true in my case. I have very few books in digital that I don't also own real copies of. The ones I don't like, I just delete.

[piranha]

it doesn't really say anything about print versus digital, just of pirated versus non-pirated. while the latter probably skews towards print, they're not identical (people pirate printed books too; always have).

i have a lot of books in digital now that i don't have in print -- heck, i buy a lot of books these days that don't even have print equivalents. i am coming to prefer the format for all but art books and large graphic novels. it's so much easier to store and search, and i don't really have a spiritual connection to holding a print book in my hands unless it's a really ancient or particularly beautiful edition; the average paperback does nothing for me.

what this graph confirms is anecdotal evidence i have that after a book is pirated, more people are exposed to it and more people buy it. though whether that holds is questionable -- this graph doesn't have reliable data for that.