Robert J. Sawyer

Hugo and Nebula Award-Winning Science Fiction Writer

More on online audiences

by Rob - July 22nd, 2006.
Filed under: Uncategorized.

Evo Terra answers some of the questions I raised in this posting, but not as a comment to that post; rather, his response is in his comments to an earlier posting by me, in which he says: “6 of the 58 titles [at] have (or have had) over 2000 subscribers. 17 are over 1000.”

So, just to be clear, about 90% of the titles at (52 out of 58) DON’T have “thousands of subscribers” and just 10% do — and we still don’t know of that 10% how many are SF. But the maximum number of data points we’re talking about, if every single one of those works ARE indeed SF, is just six.

But then Evo goes on to say that most of these are “works in progress.” In other words, apparently substantially fewer than six (and maybe even as few as zero) works in all genres at have had “thousands of subscribers” (minimum: 2,000) for AN ENTIRE BOOK.

That’s an awfully small number of data points (somewhere between zero to six) to base the claim that online serialization is “enabling many authors who have chosen the serialized route to find audiences in the thousands.” And it’s ever slimmer evidence in the case of science fiction, which was what was under discussion here, since we don’t know the proportion of that zero to six data points that are SF.

But what really interests me is how this all relates to actually putting money into the author’s pocket. Our analysis of Cory’s figures (in my own comment to this post) suggest that between, oh, say 0.5% and 2%, of those who download a freebie will buy the finished product.

My own Fictionwise numbers, reported in this blog post, suggest the ratio of those who will grab a freebie to those who will pay is in the 0.5% to 1% range.

If you have 650,000 downloads as the very first, high-profile giveaway experiment did (Cory’s first novel), and, as far as we know, no book published under this model has ever equaled since, that results in between 3,250 and 13,000 copies sold exclusively because of online exposure

[Digression: And I very much doubt the 2% or 13,000-copies figure. I also doubt that it was just downloading, and not all Cory’s other online activities, that generated interest in his book: of all the copies sold (a number Cory doesn’t divulge) of Down and Out in the Magic Kingdom, some sold because of the efforts of Tor and booksellers or as a result of reviews, some sold because of Cory’s public profile, and some sold because people read and liked the freely available text — but it’s folly to attribute all, and I’d say it’s folly to attribute even most, of the sales to the free downloading, because that implies that Cory’s book would have done quite poorly without this. There were lots of other Tor first science-fiction novels that year whose books sold well enough in hardcover to justify their release in mass-market paperback, and, even with giving away two-thirds of a million copies online, Cory’s book apparently didn’t do as well as those titles. Do the evangelists for this really want to say that Cory’s book would have sold extremely poorly had it not been given away online? If so, why? End digression.]

Anyway, if you have just 2,000 subscribers/downloaders (the lucky six at, the same percentages — 0.5% to 2% willing to pay for what they got for free — suggest that between ten and forty copies might actually sell if a version were offered commercially.

On the Dragon Page, Evo notes that he believes most first novels in the SF field sell only 5,000 copies; I suspect that’s right (and is certainly the right order of magnitude) for novels that don’t go into mass-market paperback; many average ones that do go into mass-market (and many first novels do, of course) might do at least twice as well as that in combined hardcover/mass sales.

But what Evo doesn’t mention is that the average advance for a first novel in this genre from the major publishers is US$5,000 (with, I’d say, 80% of all first science-fiction novels published by the majors falling between $2,500 and $7,500 advances). That’s not to be sneezed at, and if one’s implying “Hey, a few thousand free online downloads here, a few thousands print copies in bookstores there — what’s the difference?”, the difference is clear: thousands of dollars in the writer’s pocket.

Leave a Reply