I don't know the least thing about these people, but my rudimentary math and reading skills tell me a 300 page novel will cost about $500 if you hand them a paper copy of the book, say, sure, guillotine that puppy, and ask for a file that can be transmitted to Amazon for placement in the kindle store. My sense-of-the-world suggests that you aren't likely to get this done well for much less and you might have to pay more to get it done well. If you do it yourself and it takes you over a week, you'll be paying yourself less than $10 an hour to beat the deal.
I found this pbook/ebook, published by Perseus Book Group.
I know zippo about it, beyond that it looks like genre fiction and is published by Perseus. Whoops! It's 400 pages. Let's call it $600.
Based on this:
We know this:
"Also indicated in one client notification is that Amazon has also increased the ebook conversion fee they charge publishers who provide only a print book or PDF and want Amazon to produce the book file. That charge is now 8 percent of digital list price, up from 5 percent previously."
And we know that that client is EITHER Perseus or National Book Network. And we really don't care; I just wanted a specific example. I have no mortal clue who did the conversion work on Ms. Parker's novel, but for the purposes of this example, let's start by pretending that it was done prior to the contract renegotiation by Amazon for Perseus. How many copies would it have to sell before Ms. Parker would start feeling like someone should have forked over the $600 instead of paying Amazon 5% of digital list price (in this case, $7.99) to do the work.
Call it $.40/copy for the 5%. The breakeven is at around 1500 copies. Up until Amazon moves 1500 copies of _this particular book_, Perseus and/or Ms. Parker got a screaming deal. After that, they are getting progressively more and more screwed (unless there's some kind of sunsetting clause). If _The Dark of Day_ were to sell, say, 10,000 copies over the next couple of years, that would be roughly equivalent to paying $4000 to do the conversion.
Now let's pretend that the digital conversion was done by Amazon for Perseus and after the new contract takes effect (almost certainly not true), at $.64/copy for the 8%. Breakeven now occurs at under 1000 copies. At 10,000 copies, it would be like paying someone $6400 to do the conversion.
If you don't wind up selling a thousand copies, then you sure as hell want Amazon to do your conversion for you, and you honestly probably don't even care too much whether you pay $.40/copy or $.64/copy. But if you sell tens of thousands of copies, then you _really_ don't want to be paying Amazon to do your conversion for you.
There is another way of thinking about this. Amazon managed to get -- maybe not from Perseus, but from Perseus or National Book Network -- an agreement to pay the _same_ royalty to Amazon for doing the conversion (from print or pdf) that the author is probably getting.
There are some other things to think about. Maybe Amazon is doing such a high quality of work that it's _worth_ that much money. (Do you think that?) Maybe eBook Architects does a truly awful job, and while Amazon's isn't worth thousands of dollars, it is enough better to make a real difference and you'd have to find someone equivalent to them (or better) and see what that cost worked out to find the correct breakeven. And of course it takes time to do the shopping around and time is money and blah blah bleeping blah, but _isn't that exactly what a publisher ought to be doing_? If not hiring the people to do the conversion, then identifying cost-competitive, good-enough quality companies to outsource to? I get that the underlying publishers are too small to do this themselves and that's why they work with National Book Network or Perseus Book Group and they're supposed to get the services they need but can't manage on their own through the sort of a publisher sort of a distributor setup. Perhaps this is purely a transitional solution? Or maybe it's only used in cases where the book isn't expected to sell, but a presence is desired?
Maybe Amazon is worth 8% in perpetuity for doing the conversion. It sure is if you're not ever going to sell very many copies and clearly, that is a component in Amazon's decision to raise it from 5% -- they could surely tell they were losing a bundle on that service, right?
ETA: I do recognize that the percentage is of the digital list price, but that the actual ebooks are sold discounted, thus making the percentage paid to conversion higher on the "actual" gross. In the case I picked, the book was being discounted to $6.39, thus, 5% = 6.25% and 8% = about 10%. I'm a little unclear on whether author royalties come off the list price or the sales price.
I absolutely expect someone to chime in saying that $500 or $600 as the case may be is a criminally low price to charge and they can't possibly be doing a decent job on the conversion.