I see a major flaw here that I have to point out.
The pitch to the publishers was absolutely a fabrication of true readership of the fan fiction, that is fair call. But to say that any book with the same amount of marketing would reach the same level of appeal is short sighted.
Simply, the book had the X-factor. The story was provocative, sensual, shocking and ultimately all elements contributed to it's appeal. People talked after it was published, people talked about how it effected them and that raised awareness with the millions of people who were ultimately affected (some of which have never actually read the story, like myself). I live in Australia and from my perspective the book was never marketed as "read by millions online first as a fan fiction and now published as a big, rock-hard book".
It was the content of the story and how that resonated within society which ultimately lead to it being so huge. The X-Factor, you see it in all facets of the entertainment industry niches (I know I work in the industry). You may even find that if 50 shades of gray were released 10 years ago that it would never have gained such acclaim at all because we as a society may not have been so engaged with it's content.
The official story: "Fifty Shades of Grey" was a Twilight fan-fiction that had over two million downloads online. The publishing giant Vintage Press saw that number and realized there was a huge, previously-unrealized demand for stories like this. They filed off the Twilight serial numbers, put it in print, marketed it like hell, and now it's sold 60 million copies.
The reality is quite different.
I spoke by email to Anne Jamison, author of "fic: Why Fan-Fiction is Taking Over the World", and the person who originally reported the "over two million hits" that has been widely reported as "over two million downloads". The number two million was much too large to be possible given the size of the fandom, so I asked her about it. She replied,
Fan-fiction is published one chapter at a time. "Fifty Shades of Grey" has 26 chapters, but when it was originally published on fanfiction.net as "Master of the Universe", it had over 100 chapters. Let's say 120.
The number of hits a person generates while reading is determined by how they read it. fanfiction.net adds 1 hit every time any page of the story is reloaded. If you go to chapter 1 and read all the way through to chapter 120 in one sitting, that's 120 hits. If you log in, see it updated, go to chapter 1, and then go from there to the new chapter, that's at least 239 hits to read the book. If you refresh the page, that's another hit. (I just verified this myself by refreshing one chapter of one story of mine 3 times on fanfiction.net, checking the stats before and after.) If you read half of one chapter one day, and log in again and finish it the next, that's at least 2 hits. If you leave it in an open tab on your computer, that's 1 hit every time you open your browser. If you reread the story, the hits double. If you click on the story each day to see if it's updated, hits go way up.
The number of people who finish a multi-chapter fan-fiction is, surprisingly, almost always 40-60% of the number who clicked on the first chapter, with the very best reaching 60%, and the misspelled, grammar-free, plot-free, alphabet-soup-vomit of ten-year-olds retaining about 40%. I've checked this on a large number of stories on fimfiction.net, which records readers per chapter based on username and so avoids double-counting. The quality of a story has very little impact on whether someone who started reading it will finish or not.1
So two million hits on a 120-chapter story means a theoretical maximum of 2000000 / 121 = 16,529 readers finished it, assuming half of all readers quit after chapter 1. More likely, given re-readings, users who always go in through chapter 1, users who quit halfway through, browser refreshes, etc., perhaps 4,000 readers finished it. That would be about as many as finished a pretty popular story on fimfiction.net. The Twilight fandom had a larger fan base, so I don't find that number at all impressive.
So what actually happened was that a moderately-popular fanfiction that had been read by a few thousand people was reported on in a way that misled publishers into thinking that it had millions of readers, when really, it just had an unusually large number of chapters. They put a major marketing campaign behind it. And since 40% of readers will finish anything, absolutely anything, that they have started reading, they sold millions of copies. Just as they would have with almost any book they'd marketed as heavily.
1. This percentage range applies only to stories found by fans through the site itself. The exceptions are not exceptionally good stories, but astonishingly bad stories--89% of readers finished this piece of crap. I think this is because people aren't looking for good stories, they're looking for the sort of thing they want to read. Some people want to read very bad stories, and such stories are easily identified from their descriptions.