A self-publishing journey

I was inspired to write this post after fellow writer Katie M John chronicled her voyage into indie publishing on her website.

Like Katie, I'm an indie author, which is a trendy way of saying that I'm independent. I self-publish. I don't have the backing of an agent or publisher – it's all down to me, for better or worse.

The decision to work in this way isn't ideological. I don't have anything against either publishers or agents. Quite the opposite, I'd love to have the backing of a publisher who believes in my work. I'm indie because I'm pragmatic: it gets the job done. Rather than spend potentially years seeking a publisher, I'm out there, selling my work, generating a readership.

Of course, this means that it isn't just the writing that's down to me. Everything a publisher would do, I have to either do or organise.

Like Katie, I didn't want the fact that I'm indie to be an excuse in any way whatsoever. I want every aspect of my work to stand up alongside traditionally published works: not just the writing, but the cover design, printing, layout – everything.

I was fascinated to read Katie's account of her experiences and she encouraged me to share mine – so here they are. Like Katie's post, this is a long one.

I write using Scrivener. I've blogged about this before, but quite honestly I can't stop banging on about how good it is. If you write long-form, this is quite simply the best tool for the job – Word is a jack-of-all-trades, Scrivener is designed for authors, by authors. Check it out.

What this means in practical terms – or did, for my first novel The Well, is that at some point it has to come out of Scrivener into your final publishing tool. That's usually Word. In my case, I exported after the first draft was complete for two key reasons:

I was sending the manuscript to test readers, so it made sense if we all had a common reference point when discussing revisions – in other words, page numbers.
It had to go into Word at some point anyway, so that my editor/proofreader could work on it. She works on Windows; I work on a Mac – and at the time Scrivener was only available on the Mac.

I deeply regret this now. Word is so inefficient and slow at handling large documents that it severely hampered my revision process (at least, versus using Scrivener). I should have held on until the final draft. Well, it's all about learning, isn't it?

(As it happens, the new version of Scrivener directly exports to Kindle format and there's now a Windows version of Scrivener too.)

Ten revisions later and I was ready to call The Well complete. Time to publish.

For me, publish meant both print and paper. It's worth knowing that my strategy really is to produce e-books. But not everyone has a Kindle or other e-reader – I decided that a print version was essential for marketing purposes, if nothing else. This meant two publishing processes.

I'm fortunate that I already possess the publishing skills required. I've been a graphic designer for over twenty-five years and developed websites since the Web sprang up. So, I guess I wrestled less with this aspect than many writers would – although it still proved to be both problematic and time-consuming.

Getting my manuscript out of Word and into HTML was my first nightmare. Word's claim that it can 'save as HTML' is about as accurate as me claiming I can fly (the best I can manage is to fall with grace). The resulting HTML is a garbage mishmash of unnecessary code. Those with an awareness of HTML can be shocked to learn that Word doesn't even feel compelled to close tags such as paragraph tags, it just opens them, then opens another. It also adds in a massive amount of code which does nothing except increase the file size by around 35%.

I experimented with various tools to clean up the code, but I wasn't really happy with any of the results. They improved the code, but it was still far from perfect. Since I can read HTML perfectly well, I decided to swallow hard and code it by hand. That sounds extreme, but actually it's not that big a job – the manuscript may be big but it requires very little formatting. Most of the work was in stripping out the rubbish code. The next large chunk of work was replacing UTF-8 smart quotes and the like with XHTML entities. You could argue that I didn't need to do this, but it does make the HTML more transportable. So, every instance of an open double-quote for instance would need to be changed to “ – this was really a matter of search and replace. I did the same for accented characters – in The Well one of the characters slips into Romanian, so these had to be in there. I've read a lot of Kindle books where 'real' publishers don't bother with this, so all the reader gets is a word with an incomprehensible ? or ! in the middle of it. Poor show and lazy work.

Eventually, I had a clean HTML document. The Kindle file format is really a zip file with a different extension, with HTML and XML files inside it. The Kindle format also uses its own specific tags where there's no HTML equivalent – such as to create page breaks. To create this file, I used a PC tool called Mobipocket creator (the Kindle format was originally .mobi – purchased by Amazon). Some aspects of this were a little fiddly and it did require a basic knowledge of XML to get the job right, but it wasn't too bad. I was pretty astonished to find that while there are a few tools which claim to do this kind of thing, many are really not fit for purpose. Again, next time, this won't be an issue – I can export directly from Scrivener to Kindle format. Although I work on a Mac, since I test websites I also run Windows and Linux, so using this PC-only tool wasn't an issue.

A friend of mine who's a professional illustrator, Daryl Joyce, painted the cover. I've blogged about this previously, so I won't cover that again, other than to say that while it's true you can't tell a book by its cover, people certainly buy a book partly based on its cover – so it was important that this was as good as a commercial book cover. The cover is embedded in the Kindle file as a relatively low resolution JPEG, but on the print version it looks superb.

That done, it's a relatively straightforward process to load the file onto Amazon and make it available for sale in the UK and USA (well done, Amazon).

The print version had its own challenges. I'd originally intended to use CreateSpace, since it's part of Amazon – but CreateSpace can only make titles available to Amazon in the USA and not the UK, amazingly. So, I opted for Lulu. Unlike Katie, I didn't want to invest in inventory – the print version for me is mainly a marketing tool. In fact, I literally make pennies from print copies sold on Amazon after Lulu and Amazon have taken their cut. I'm OK with that – I wanted to make it as affordable as I could. Print on demand means higher unit costs, so I was trying to offset that.

Because I'd originally intended to use both Lulu and CreateSpace, I'd opted for the smallest common book size, which was A5. As it happens, this is also quite economical – costing not much different than a pocket book size, but able to have more words per page. As I was about to learn, the number of words per page has a drastic impact on the cost of the final book.

I created a layout in Adobe InDesign, and flowed into it my 116,500 words. The resulting book would have retailed at a commercially unrealistic £16.50. Clearly, I needed to be more brutal with the book layout. I pulled down a bunch of paperbacks from my shelves and measured the font sizes and margins. I reworked the book to something more realistic – allowing for a £7.99 retail. My first printed proof showed me that I was still a little off – the gutter was too tight, so the book would have to be forced open too much to read it. I adjusted the gutter by removing space from the outer margins. (It's worth noting that quite a few novels are 60,000-80,000 words, which would have really brought the costs down. Perhaps I'll consider that in the future but The Well is a pretty tight fast-paced read, I couldn't have lost 20,000 words easily.)

I chose a typeface which had professional ligatures, so that I would have nice-looking fl and fi characters for instance. I had looked at typefaces which were used in traditional publishing such as Bembo, but quickly realised that these had been designed in the days of letterpress printing where there is ink gain on impression. When printing litho, there is usually a little loss. I opted for Warnock Pro, which is nicely economical in terms of width and had a decent x-height so it is very readable.

My comparisons with commercially published books showed that many would reduce the typeface further than I had and also have tighter margins. If I'd followed the same route I could possibly have saved a little more in terms of pages and costs, but these looked too tight to me. If I was working with a publisher, this wouldn't have been my call, of course.

It took a few proof copies before I was totally happy. Since Lulu gives you a free ISBN and can push from their systems onto Amazon, that gave me everything I needed. Click – publish. It was available on Lulu pretty much right away and then appeared on Amazon several weeks later.

There was a bit of faffing around with Amazon to get them to combine my print and Kindle editions – and the reviews – but they were very responsive, actually, and always helpful. The same is true of Lulu, though they can't combine reviews of both my print and ePub editions. I created an ePub edition of The Well so that it would feed onto the iBookstore, but it's yet to appear. (I'm hoping this isn't because I sent a shirty e-mail to Steve Jobs criticising iBooks, and the smooth experience of being a Kindle author versus getting a book on the iBookstore.)

So, even for an old publishing hand, there were quite a few bumps in the road and a lot to learn. I've yet to get The Well onto the Sony Reader and Nook, which I guess I could really do with doing. (Any indie authors who've done this – I'd welcome hearing your stories.)

The next time around should be a lot smoother, partly because of what I've learned, partly because Scrivener now supports Kindle export, partly because I can hopefully keep the next book in Scrivener for more drafts and hopefully because Amazon will further improve their processes.

I'm not one to say 'you don't need a publisher'. Publishers offer a lot of value and take from your hands some time-consuming tasks. But it can be done without them – you've just got to be prepared to roll up your sleeves and expend some sweat!

Filed under: Writing