Lost in Amazon 5: How to NOT Look Like Crap

I may owe some people an apology. It seems I scared a few of them with a comment I wrote on a Stephan Metral buzz over on beBee.

I wrote about the problems I was having getting an eBook to actually look like a book. I concluded the best way to go was to:

• Create an ePub format of the eBook
• Smash it up into its component files
• Modify the HTML with extensive use of GREP (RegEx)
• Stitch it all back together.
• Upload the repaired package to Amazon
• Hope/pray that it works.

I also mentioned I wasn't looking forward to trying that.

Some people got frightened. I don't blame them. I was pretty frightened too.  I really didn't want to do all that just to put out an eBook 

This installment in the series will go over the methods I used to create the files that make up an eBook. I'll go over which methods worked and which didn't work well enough to bother with going forward.

I'll go over the pros and cons of each method. I'll also include an estimate of what you should expect to pay to have someone do it for you, or an estimated cost in software and time if you choose the DIY route.

A review of what this series is all about

This installment is the fifth in the series where I try to crack the Amazon code.  That's fitting, no? As in, "I plead the 5th," or, "This process is driving me to drink. Pass that fifth of bourbon!"

I digress.

We're using an "eBook as widget" philosophy.

Amazon is a big, busy market. It should be possible to earn a reasonable, semi-passive income writing eBooks. At least, that's the idea. It has been said that "should" is the saddest word in any language.

I'm starting to agree.

I promised to report back on successes and failures both. So far, there have been damned few successes.

eBooks generally don't carry big price tags. Any extra work involved in getting the finished piece done bites deeply into the economics behind the idea. Busting up and rebuilding an ePub package definitely qualifies as "extra work." 

eBooks are, at their core, just HTML. Yeah, it's the stuff you are looking at right now. You would think it would be a simple thing to get one to look like something other than crap.

It isn't.

Try #1 - Amazon says it's easy. Upload a Word file and BOOM you're done.

Not quite. 

Yes, you will have an eBook. No, I wouldn't consider it saleable. It might be good enough for what you intend to accomplish. It might not be. 

It probably won't be.

You owe your buyers a decent product. Imagine this. You order ravioli in an Italian restaurant. You see them open a can of Chef Boy-r-Dee and dump it on your plate. How would you feel?

This method, done as is, is very similar to that.

The result is butt-ugly and typographically yucky. To repeat, it could be good enough. I'm not one to be satisfied wth "good enough." I believe an eBook should look like a printed book.

More complex, more typographically rich layouts like a non-fiction book are much less likely to be adequate. It is possible to do with some reworking. 

I don't know him but Aaron Shepard wrote 3 books that will help. I bought all three. The first two will help a lot. If you need the third, you probably shouldn't be using this method unless you have no alternative.

From Word to Kindle discusses how to set up your source Word file. What HTML fixes for Kindle discusses is self-explanatory. The third, Pictures on Kindle, is also well-done. I just believe that if you have so many images in your book that you need this one, you probably shouldn't be using this method in the first place.

Aaron's books are proof that it's possible to create decent quality non-fiction books using this method.

Estimated Cost: Assuming you have access to Word, this method will set you back a whopping $8 to $12 for Aaron's books.

If your book is fiction and is only a title page followed by rivers of text, this method, using Aaron's fixes, could be enough. If your book is non-fiction, using Aaron's fixes might make it good enough.

I'm more a non-fiction type of guy. Complex typography actually makes things easier to read. I also think that "good enough" rarely is.

I kept investigating.

Try #2 - Amazon has a plug-in for InDesign. Woohoo!!!

My dreams were answered! InDesign is to page layout what Photoshop is to image manipulation. I figured I could import my Word file into InDesign. Then I could use InDesign's super layout abilities to create an eBook that looks just like a real book. 

I'd be done in no time flat!

Silly me.

The plug-in is buggy as all get-out. When it worked, most of the nice layout touches were ignored. It's a complete waste of time. Yes, it will make a nicer looking book than the Word method, but, again, only for straight-text fiction.

Estimated Cost: Why bother calculating it? See Try #3 instead.

Try #3 - ePub upload

Once I had my test book loaded into InDesign, I tried other methods. Amazon accepts uploads of ePubs. Also, ePub is the format preferred by pretty much every sales platform other than Amazon. (Yes, there really are others.)

This is the way to go for text intensive fiction books.

The initial work is loaded into InDesign. Then, it gets formatted nicely before exporting to ePub format. That ePub can now be uploaded anywhere.

Estimated Cost: InDesign is not cheap. If I remember correctly, Adobe Creative Cloud is about U$50 a month. Also, the learning curve is very steep. It's not worth buying unless you plan to crank out book after book. 

It should take an hour or two to do. Much depends on how well the initial document was formatted. If you get it done, I think anything under U$150 is fair. Most service providers will provide a framework or template for your book.

Try #4 - Thank you, Robert Bacal

It was about this time that I resigned myself to the nightmare of busting up an ePub, performing major surgery on its guts, re-stitching it all back together, and hoping the "patient" survives. 

I really wasn't looking forward to it.

Then I remembered a comment Robert Bacal once made. It was something about print-replica eBooks. Darn if I could find that comment, though. That memory led to the germ of an idea, which led to a ton of research, which led to an answer.

I passed over that answer four times.

Amazon has something they call Kindle Textbook Creator. 

What is a non-fiction book if not a textbook for people out of school? 

I dug into it.

The more I did, the more I dug it!

Kindle Textbook Creator is not without its drawbacks. 

It only works on smartphones, tablets, computers, and the newer readers. Anybody else will not even find the book in a search. I'm okay with that, especially when the alternative is the whole exploding ePub thing. 

On the plus side, all these things also support color, video/audio, and pinch-to-zoom.

The resulting eBook is not reflowable. To make a similar eBook that is reflowable and will work on any Kindle, you need to go the exploding ePub route. I doubt the extra potential distribution can cover the associated cost in time and aggravation.

You can add graphs, charts, images, code snippets, and formulae to your Kindle Textbook. No problem there.

You can add rich media like video and audio. You can add them as embedded media or as links. Heck, you can even add complete Powerpoint slide decks, narrated or not. 

Make sure you keep an eye on file size! Don't force yourself into 30% royalty purgatory. The 70% royalty system carries a delivery fee of $0.15 per megabyte. Don't let that fee eat up your extra royalties. 

Rule of thumb: Keep your eBook under 5Mb.

How KTC works

Kindle Textbook Creator is a free desktop application. Free is usually a good price. The InDesign plug-in is also free. For the plug-in, free is pricey.

Kindle Textbook Creator is intuitive to use. I think my favorite non-techies, Lisa Gallagher and Susan Rooks, can make a go of it.

KTC exports to KPF format. That's Kindle Package Format, not mobi or kd8. It can only be used on Kindle Direct Publishing.

The source document is a PDF file, but you can't just grab any old PDF and use it. Most PDFs are meant to be printed on letter-sized paper. That will translate directly to a screen size. Imagine what a letter-sized page would look like shrunk down to fit your phone.

Everything would be 25% of its initial size.

Your readers will hate you.

Your readers' optometrists will love you.

The file needs to be reworked to a much larger font size. Images and charts may also need a rework.

My first try was at a font size of 14pts for regular text, using the Arial font. That just annoyed me. Let's just say that pinch-to-zoom is nice, but not if you have to do it on every page, then pan around to read the frigging thing.

My second try was at 16pts. That was better but still not quite right. The third try at 18pts did the trick. The original document was at 11pts if anyone's curious.

I can see that some people with smaller smartphones may find even 18pts annoying. A "large-type" version is a definite idea, even if it requires a rework. I'll probably look at a 22pts size and see what happens.

Navigation is not by hyperlinks through a Table of Contents but by PDF bookmarks. Make sure you add them, and any hyperlinks to your initial work before you export to PDF.

I'm undecided if a formal Table of Contents is desirable. It isn't necessary. I wonder if people expect one. It's always a good idea to meet expectations. I'd appreciate your opinion. Just leave it in a comment.

Word can add bookmarks and hyperlinks. Word can also export to PDF. 

I can cook a steak with a blow torch, but a grill works much better. So does InDesign for PDFs.

Besides, once in InDesign, I can do quintuple duty. Apart from the PDF, I can make another PDF version to sell on my website (another, "Thank you," to Robert Bacal). I can export it to ePub. With a little more work, I can easily export a second "large-type" version to Kindle Textbook Creator, then use the same file to make a large-type version ePub.

Images may be an issue if you start from a print-ready PDF. Print images are in CMYK (four-color process.) Electronic formats, including eBooks, need them to be in RGB (red-green-blue). You will need to convert them or have them converted. Amazon says they convert them automatically. I wouldn't trust them on that. Charts and graphs should be okay. Pictures may come out moiréd.

This method is not suitable for text-intensive fiction. The richer, the more illustrated (within limits), the more typographically complex the work, the better suited this method is. For straight text, losing reflow ability is neither an option nor desirable.

Estimated cost: If you have Word or anything similar and you set your file up properly this method should carry no cost at all. 

Having it done by a service will vary a great deal depending on your source file. Again, most service providers will supply a template and/or instructions. Count on anywhere from U$100 to $500 for a modified PDF, a Large-Type version PDF, and a couple of ePubs (also modified and large-type). You'll have all bases covered.

Closing

The first test book will be published tomorrow. It's a how-to for implementing PayPal's Instant Payment Notification and/or Payment Data Transfer systems for e-commerce sites.

It's the book that I wish existed when I started trying to get those beasts to play nice for myTweetPack.com. PayPal is a great company with a great service. Its documentation is considerably less great.

I'll add the link to it once it's published. If anyone wants to read an advance copy, leave your email in a comment or message me directly. All I ask in return is that you leave a review if you liked it.

I'll also add this post to an updated version of The Complete Index to The Lost in Amazon Series post.

The second test book will follow soon. It too will use Kindle Textbook Creator. That one will explain the diet I created from my culinary school training to lose 60 pounds in 9 months eating good, satisfying, yummy food. Even my not-particularly-overweight wife lost weight with it. She complained the whole while saying, "This is a diet?!!? It's just food!" 

I call it The Chef's UnDiet.

Last, I'll try my hand at fiction. (Be afraid. Be very afraid.)  LOL

The next step will be to use those test books to figure out a promotion and marketing plan. 

Cheers