Article placeholder image
Training the beast

Training the beast

A slightly niche topic this week, but an important one.

On the whole, new writers think, “I have to make sales. It doesn’t matter who I’m selling to. I just need to sell as much as I can, and the more I sell, the more booksellers will love me, and the more publishers will love me, and the better my chances of being asked to write another book.”

That sounds terribly logical and, in bricksandmortarland, it is logical. Sales are sales. A supermarket doesn’t know or care whether Customer X is or is not the ‘right’ customer for the book they’ve just bought. If it turns out that your gran went shopping without her glasses and accidentally bought your sweet historical romance instead of the blood and guts Viking-monster-horror stuff she normally buys – well, that’s her tough luck. She discards the book unread. The superstore has its money. No one (except you) cares much about your gran.



Amazon doesn’t work like that or think like that, and you need to be careful.

So let’s just say that you have a sweet historical romance to sell. Very imaginatively, you have called your novel My Sweet Historical Romance. The cover depicts a maiden dressed in white lace looking shyly up at the duke who will (by chapter 33) be thoroughly smitten with her.

Let’s also say that you are (I know you are) particularly kind to your gran and her circle of Viking-monster-horror-loving friends.

You ask your gran if she wouldn’t mind buying your new book. She’s happy to oblige. Her friends are also happy to oblige. You get a couple of dozen sales early on, when your book is newly launched.

Not bad, huh? I mean, two dozen is only two dozen, but Rome wasn’t built in a day, a journey of a thousand miles begins with the first step, and a ragbag of other assorted cliches to boot.

Except Amazon has a new product to offer and it has to figure out who might want it.

At first (judging from the cover, the title, your book blurb and all that), Amazon guessed that your book would mostly appeal to people who liked sweet historical romance and shyly smiling maidens and all that. But then a wave of Viking-monster types charged in and bought the book. So Amazon tries offering it to other people with those reading preferences.

At that point, one of two things happens.

Either, loads of people are shown the book and don’t buy it, so Amazon thinks the book must be rubbish and stops marketing it.

Or, people do buy the book, find it different to what they were expecting and to what they usually read, and stop reading the book partway through – and perhaps leave lacklustre reviews to boot.

These outcomes are both catastrophic.

The Amazon-stops-marketing-your-book option is bad, because you’ve just lost the services of the biggest and most sophisticated book retailer on the planet.

The lacklustre-response option is equally bad, however, because Amazon knows (via its huge Kindle-reading base) how much of an ebook gets read. If your book doesn’t get finished, Amazon will prefer to market what it sees as better products. Further, those lacklustre reviews are going to be a stone around your neck for years to come.

Either way, getting the wrong readers into your fiction early on will cause lasting (and very hard-to-reverse) damage to your sales.

If all this sounds a tad theoretical, then stay tuned.

Plenty of (mostly indie) authors aim to make money by using ads to direct traffic from Facebook to Amazon. What could possibly go wrong, right? For relatively small amounts of money, you can fish in the largest pool of users in the world and send them to your very own page in the world’s largest bookstore.

But unless you are careful to get Facebook sending the right readers your way, you’re going to end up sending the wrong ones. And, OK, you probably won’t find that you are sending Viking-monster-horror readers to your My Sweet Historical Romance bookpage, but you might find that you are sending (say) lovers of billionaire romance to your book page. Or lovers of raunchy romance. Or other readers in nearby but definitely different niches.

Any such misalignment of traffic and product will be just as injurious as the Viking-monster-horror example I started with.

You’ll get weak conversions, poor reviews and people failing to finish your book.

This email was sparked by a message from Dave Chesson at Kindlepreneur, who was running a profitable Amazon ad. (Or Faceboook ad; I forget which.) At one level, the arithmetic was simple. Chesson was spending X. The ads were generating X-plus-something. Everyone a winner, right?

Except that the ad was badly targeted. He noticed he was getting poor reviews – and reviews from people who clearly were not his target reader.

It’s not that easy creating an instantly profitable ad, but Chesson had done it. Yet he saw that the ad was doing him long-term injury, so he killed the ad. Protecting the quality of the book’s traffic was more important than making a few dollars of profit in those early days and weeks.

He was absolutely right to take that step. I’d have done the exact same thing.

Now, to be fair, there’s a big qualification here – namely, that once you have a decent sales record with Amazon, the beast will essentially know your readership and a few left-field readers won’t especially impair its ability or willingness to market your book. After all, readers are eclectic, and Amazon knows perfectly well that sometimes Viking-horror readers do also like a shyly smiling maiden or two. (And not just to sacrifice.)

But you have to approach the Amazon sales process in order, always. Train the beast carefully. Then feed it.

You’ll know when you have it nicely trained: your “customers who read this book also read” list will look like a nice collection of comparable authors. The sponsored ads (“Based on your recent views”) will also look, for the most part, like a logical collection.

This advice comes in large flaming letters for indie authors. For trad authors, who just don’t have much control over what their publishers choose to do, it makes less difference. But even there, just remember that bad sales are worse than no sales, especially early on.

Here endeth the lesson.

Feedback Friday

I’m doing a live feedback event next week, so your task this week is really simple. Premium Members can register for it here. 

I want, please, 250 words (max) that you really like. Also, title and genre.

I’m going to give live feedback on this stuff next week, so if you don’t want your work torn to shreds in front of a baying mob, please mark it: NO LIVE FEEDBACK.

(Truth is, I only pick work I already quite like and I’m never that mean. But if you don’t want the live experience, then please just tell me so.)

That’s it from me. Post yours here. I’m off to slay a sea-monster and plunder a couple of Lincolnshire villages.

Til soon.


Related Articles