Self-publishers fail to earn a fortune shock!

The Taleist today published the results of its survey into self-published authors' earnings and their habits. They asked 1007 authors 61 questions, which mkes it undoubtedly an important event. The headline figure is that self-published authors earn $10,000 a year from their books on average (£6375) whilst half earn under $500. The media is already buzzing with the seeming negativity of this - though in reality it's hardly quantum leaps off the mark of the Society of Authors figures for published authors (£16k mean £4k median).

There are some fairly bland observations - people who write romance earn most by genre, people who write literary fiction don't earn much, you do better if you hire an editor, but the report will no doubt go viral, and for me that is worrying because - whilst it can hardly but wear its unscientific colours on its sleeve, it seems to me much more a "how to" guide than a state of play survey, and that means authors are going to jump on it and maybe overlook many of its gaping holes - by not being exhaustive and rigorous, it's impossible to use the information in this report as a definitive how-to, because it fails to analyze what may be other central contributing factors to the bestsellers' success (in addition, that is, to, DUH, how good the book is as a representative of its type, which I've not seen mentioned in anny of teh coverage so far!!!!)

In addition to the survey size, I can't find the original post asking for respondents on the Taliest website, but the fact the actual report being charged-for at £3.30 a pop raises questions - I'd like to know whether it was made clear to participants if they were undertaling the survey as part of a commercial enterprise, for example. The purpose seems to have been mainly to produce a "secrets of bestselling authors" document rather than a state of play one, as I've mentioned. No problem in that, but this angle and the commercial aspect combined (one can't help but compare it to John Locke's "How I sold a million" only "how people sell millions") need to be taken into account before people simply red the results off the page as if they were instructions.

That said, the conclusions seem sound, if somewhat obvious - romance sells best, you sell more when you use an editor and cover designer. As how-to goes, though, authors would probably be better served by the 99 cents for Locke's book. What we really need for a definitive "secrets of success" guide is a survey that goes right into the detail of metadata, for example. At one of the events at London Book Fair, an Amazon representative said the single best piece of advice he'd give an author trying to make sales on Amazon was get their metadata right. It would be great to take some cross sections within genres and see some segmented data based on review stats, metadata, price, inclusion in charts, promotional sites used (one thing authors desperately need research on is just how effective in the long run paid-for promotions on sites like Kindle Nation Daily are, with control groups and longitudinal data over several years cross-compared with the dates of other major promotions such as Amazon newsletters and sales containing books in the same genre so authors can see the effect of timing their promotion).

One thing that sounds very familiar from listening to self-publishing debates but seems to me to be utterly contentless:

It shouldn't have surprised me that 75% of the royalty pie is going to 10% of authors: that's life in many industries. If I'm being honest, though, I'd hoped self-publishing might be a bit more democratic

What is that supposed to mean? That they hoped in a self-publishing world everyone would sell exactly the same amount of books? (never going to happen) That they hoped everyone would appear in the same marketplace? (they do) That they hoped sales would be less affected by variables reflective of a profesisonal attitude like getting in a cover designer and editor? (maybe they hoped it but I'm sure readers didn't) That they hoped sales would be more bunched with less outlier-skewing? (they are but the survey is too small and self-selecting to reflect that)

This report is interesting and I would advise athors to read the coverage with interest. Worth £3.30? Maybe, but only if you go in with eyes wide open about what you can and can't expect to get out of it.

And a note of scepticism - many of the questions I want to ask, even ones I've raised here, would no doubt be answered by thereport's authors with the line if you buy the report, you'll find out" - and that most definitely is not the mark of research done for the sake of information.