A/B Testing for Independent Authors

A/B split testing is one of those business school concepts all writers will find easy to understand, but might not have been exposed to in school or during their “day job” careers. It works like this:

Step One: Create two iterations of the same product, with one or more (ideally one at a time) minor differences

Step Two: Put them for sale under as similar conditions as you can manage.

Step Three: Watch which one sells better.

Step Four: Move forward with just the more successful version.

It’s all well and good to think one thing will sell better than another, or draw more subscribers to a mailing list, or prompt more visits to your website, but with A/B testing, you know for sure.

A/B testing is a powerful tool for myriad aspects of your writing and publishing career, too many to list in full in any one blog post. Instead, we’ll look at two examples that have extreme power, and how you can apply the concept.

Testing a New Book Cover

Whoever said “you can’t judge a book by its cover” has never worked in a bookstore or a library. People judge a book almost exclusively by its cover, especially when shopping online. That means you should fine-tune your book cover as much as you can.

Some aspects of an excellent book cover include:

  • Completely free of spelling and grammar errors in the text
  • Genre-appropriate art and formatting. Look at the bestsellers in your genre for examples and inspiration
  • Attention-getting, so it stands out enough to warrant a longer look.
  • Unique but readable font, or a font used by many books in the genre.
  • Good kerning and positioning of cover copy.

For Amazon ebooks, also be certain the cover looks good in the miniature thumbnail photo for a listing and when blown up on the purchasing page.

Use those criteria to make an upgrade for the cover you have up already, then proceed with the A/B test.

Step One: Create a new ebook in your KDP dashboard, with the same interior and description but the new cover.

Step Two: Leave both books up for two to three weeks.

Step Two-and-a-Half: You can, if you want, run a Kindle Select free day or price discount. If you do so, run the same deal at the same time for both books.

Step Two-and-Three-Quarters: If you run any other promotions for the books, like ads online or review pushes, do them for both books at the same time in the same way.

Step Three: Look at the sales for both books.

Step Four: Keep the one that did better.

If you haven’t published your book yet, you can do this from the beginning by putting up two new books at once on KDP. You can also run these tests before launching your book by using two covers in a set of Facebook or Google ads, and seeing which ones generate the most clicks.

The same goes for your mailing list. Split your list in half, and send one cover to one and one to the other, and see which generates the most clicks.

Testing a New Price

Price is a funny thing when it comes to selling books. Although a dollar isn’t really a big deal in the context of most people who own a device capable of reading ebooks, it can make a surprising difference in how many books you sell.

It’s possible to test two different price points using the same basic framework described above for titles, but honestly that’s more work than necessary.

Instead, you can use a KDP Select price deal that reduces the price of your book for a few days, then offers a slightly higher price, then an even higher price, until it returns to normal. This suffers a little because different prices on different days is actually changing two things, but the simplicity of the method usually outweighs that.

When your promotion is over, find out which price generated the most revenue. You want revenue instead of sales, because selling 10 books for 10 dollars each makes more money than selling 15 for five each. Once you’ve nailed down the best price, keep your book there except for the occasional attention-getting promotion.

But What Price Should I Set for My Book?

Start by looking at the other books in your genre, especially the titles in the top ten. Look for trends there, and set the average as your highest initial price. That’s usually between $3.99 and $7.99 for most genres.

As a general rule, if your main goal is to earn royalties, keep your prices in the $2.99 to $9.99 range. On Amazon, pricing higher or lower than that window disqualifies you for the best royalty rates.

Finally, keep in mind that a price increase sometimes actually raises revenue because it implies the book is more valuable than those priced lower. This only works if your cover and description (and to a lesser extent your author presence and profile) support that assumption. On the other hand, it could also lower sales significantly.

The only way to find out is to test different prices as we describe here. Then you’ll have data – scientifically derived numbers – to tell you what the best price for your book is.

Other Things to Test

Like we said at the beginning, an author can A/B test so many aspects of their book design, sales system, and marketing. Here’s a (very) incomplete list of some other ways to use this powerful tool:

  • Write different descriptions for your book, using the same text and cover, to see which one does better
  • Test different invitations to join your newsletter, narrowing down options one by one until you find the method that works best
  • Run launches for your book using two different titles, to see which gets the best reception
  • Test different advertisements on Facebook or Google Ads