Think people don’t judge books by their covers? Think again. A book’s cover is the first thing potential readers see — whether they’re browsing online retailers’ search results, skimming their social media feeds, or perusing their BookBub emails.
A great cover design can have a major impact on book sales. For example, romance writer R.L. Mathewson went from selling five or six copies per day of her novel Playing for Keeps to over 1,000 per day by updating her cover design.
With so much at stake, why hedge your bets on one design without testing it first? Instead, use data to choose your cover design — either for a new book or a relaunch — by testing two or more variations against each other. This strategy can help both indie authors and publishing houses increase sales!
Cover elements like the image, font face, featured characters, and colors all impact the emotional reaction potential readers might have to a book. Best practices differ by genre, and like everything else in publishing, opinions of cover designs are subjective. This is why it’s so important to run your own cover design tests to find out which one will resonate most with your own unique audience!
How should the two variations differ?
If you’re redesigning an existing book’s cover, you can simply test the old version against a new version. This will give you the chance to make sure the new cover has a positive impact on sales, and avoid unintentionally hurting sales if the new cover doesn’t work as well as the original. But if you’re releasing a new book, you’ll need to create two cover designs at once.
For either scenario, what should you change in the second design? In general, there are two different strategies:
Option 1: Test drastically different designs
The two designs can be dramatically different, which can help steer you in the right direction if you’re early in the design process and unsure what will resonate with readers. Keep in mind that this may be more expensive if you’re outsourcing your cover design.
Option 2: Test incrementally different designs
If you’ve already settled on the essential look and feel of the cover, you can test individual variables in the design. This might include:
- Font face
- Font size (of the title and/or author name)
- Color scheme
- Characters included (for example: male vs. female or proper attire vs. scantily clad)
How can you test these cover designs?
There are different methods for testing multiple cover designs against each other:
- Quantitative: A/B test two designs with equal, randomly-selected audience segments. Results are based on raw data (e.g. the number of clicks) and performance results.
- Qualitative: Poll readers and ask which of two covers they prefer. This is a subjective approach because respondents see both variations.
Let’s dive into some specific testing methods available for each of these approaches!
Option 1 [quantitative]: Run A/B tests on a display advertising platform
BookBub Ads, the self-serve ads that appear at the bottom of BookBub’s daily emails, makes it easy for authors and publishers to A/B test cover designs. Simply run two ads simultaneously — each with a different version of the cover design in the creative, and otherwise identical — and compare the click-through rate (CTR) for each. The variation with the highest CTR is the winner.
Before launching a new cover for her thriller Irreparable Harm, Melissa F. Miller ran a BookBub Ads campaign to test the redesigned cover against her existing cover.
Melissa ran $10-budget campaigns for each of these variations. She opted to use CPC bidding (with a $1 CPC bid) and targeted a narrow audience of readers interested in either herself or two comparable authors AND the Thrillers category. She also targeted all retailers and regions where this book was available. And the results showed a winner — the new cover design had a 17% higher CTR!
To A/B test a book cover using BookBub Ads, create a BookBub Ads campaign where you upload your own creative. This will allow you to include a book cover image that isn’t live on retailer sites for one (or both) of the variations.
If the book is not yet available for sale, either set the click-through URL as the preorder page on retailer sites, or link to a page on your own website providing more information for interested readers.
Once you create the first ad, save it as a draft, then duplicate it by clicking Select Action > Copy Ad. This will let you carry over all of the settings from the first ad.
Everything about the second ad should be exactly the same — timing, budget, CPM or CPC bid, targeting, and click-through URL — except for the creative, which should include the new cover design. Ideally, the creative should be identical except for the book cover image. Once you create the second ad, make both ads live. Running both versions simultaneously ensures that you’re testing the impact of the design change, and not another factor like timing or day of the week.
Note that when two ads run with the same targeting and bid amount, the BookBub Ads platform will randomly choose which ad any given user within that targeted audience will see. This means that you should get a similar number of impressions for each version. However, if you’re running a CPC campaign and aren’t garnering enough impressions in total, try running a CPM campaign instead.
When the campaign is over, you can see which version had a higher CTR — that’s the winning design!
You can run similar tests across various ad platforms — such as Facebook Ads and Google Display Network — to see if the results are consistent across different audiences. See how Reedsy validated their redesigned book covers by A/B testing them on Facebook here.
Option 2 [quantitative]: Run an A/B test email campaign
Do you have a mailing list of blog subscribers or people who have signed up to receive email updates from you? If so, send them a split A/B test email to promote this book.
The emails should be identical except for the cover image — variation A should include one cover design, and variation B should include the other. Make sure to include links in this email so you can measure the CTR and determine a winner. If the book has been released, link to a retailer product page. If not, link to either the preorder page on retailer sites or a page on your own website providing more information for interested readers. The variation with the highest click-through rate will be the winner!
When Rachel Shane considered launching a new cover design for her new adult romance The Game of Love, she used Mailchimp to A/B test the redesigned version against the original cover. She found that the email including the original cover had a 35% higher click-through rate, so decided against using this new cover design.
Here are some reasonably priced email service providers that make it easy to run split A/B test email campaigns:
Option 3 [qualitative]: Use polling software
Polling software like PickFu makes it incredibly easy to test different book covers and titles. You simply need to upload two different versions of a cover, and respondents will select which version they prefer. You can get results quickly since PickFu has an established testing audience, and survey respondents need to say why they made the selection they did — this way, you’ll know they didn’t arbitrarily choose one design over the other.
Mandy Rosko used PickFu to test two different cover designs for the second book in her Dangerous Creatures paranormal romance series. You can view the results on PickFu here!
You can also read voters’ rationale on why they chose one design over the other.
PickFu isn’t free — you can purchase polls a la carte for $20, and you’ll get 50 max responses per poll. Or you can get a monthly subscription that let you run more polls to more respondents.
Option 4 [qualitative]: Run a poll on your website or blog
If you already have an established audience and are confident you’d get a high volume of responses without using polling software, you can run a poll on your own website or blog. This is a plus because you’ll get responses from the people most likely to purchase your book. However, it could take longer to get results. If you go this route, you can either use a poll plugin or rating plugin to tally results, or simply encourage visitors to comment with their choice.
Swoon Reads (a Macmillan imprint) runs a poll on their blog for each book they launch. They let respondents rate three different designs for each book, and the design team finesses the winning variation to create the final cover design. For example, they let readers choose the cover design for Tiana Smith’s Match Me If You Can — see the full blog post here.
Within this post, readers were encouraged to rate each version. The cover with the best rating would win!
If you can’t install plugins on your website, you can also run a simple poll by using free tools like PlayBuzz. PlayBuzz lets you embed a poll on a website page or blog post, and there’s no limit to the amount of responses you can get. You can also easily share this poll on social media sites to get more traction.
Option 5 [quantitative]: Stock different versions on bookstore displays
If you have print distribution, it’s also possible to test different cover designs in person! Berkley, a Penguin imprint, created dueling covers for the novel Virgin. They printed two different books with unique ISBN numbers. One of the covers was edgier than the other, and instead of speculating which design people would respond to, they let data decide.
While the edgier version of a woman appealed more to men and some booksellers, Penguin’s Facebook audience reacted better to the subdued rose V design (via a poll they ran on their Facebook page). One bookseller even opted to purchase both variations, and let the customers decide.
What are some tactics you’ve used to test your cover design? Let us know in the comments below!
Want to share this post? Here are ready-made tweets!
Click to tweet: Want to increase book sales? Try a new cover design! Here are several easy ways to test multiple designs and determine a winner. http://bit.ly/abtestco
Click to tweet: A great cover design can have a major impact on book sales, so consider testing yours before a launch or re-release! Here are several easy ways to tell which design will sell more copies. http://bit.ly/abtestco