I have a wonderful guest post today from author Nicholas Erik – a skilled marketer with a keen analytical mind, who will show you how to analyze the profitability of your paid ads, so you can sell more books. It’s easily the longest post we’ve had here – over 4,000 words – but it’s worth settling in for this excellent, comprehensive guide which will have something useful for all self-publishers, from beginners to experts. Here’s Nick with more:
As the title might suggest, this guide is all about how to analyze the profitability of your paid ads to help you sell more books. The step-by-step system outlined within is effective for both promo sites and pay-per-click (PPC) platforms such as Facebook Ads, Amazon Advertising, and BookBub.
If you follow the steps in this guide, you’ll have solid numbers on which to base your advertising decisions—instead of making guesses regarding their profitability.
I want to be clear: paid advertising—particularly PPC—has a steep learning curve. Developing your own skills demands extensive testing and practice. It is likely, in the beginning, that your ads will be unprofitable.
If your budget is limited, do not spend money you don’t have. Start small, and continually reinvest a portion (or all) of the profits back into your publishing business. Those with no budget can still apply many of the concepts to increase the effectiveness of their free marketing efforts.
Since losing money is possible, many authors are hesitant to advertise and view it as risky. Done correctly, however, paid advertising is not risky and is instead a reliable profit generator.
In fact, increased competition (over 7 million books and counting on the Kindle Store) and vastly decreased organic visibility make it difficult—though not impossible—for authors to be successful in 2018 without spending money on ads.
Successfully analyzing your ads involves calculating three simple metrics: cost per download, sellthrough, and revenue per download. But before we begin with the math, it’s important to lay out seven principles critical for profitable advertising.
How To Use This Guide
There is a lot of information here, but if you take it slow I hope you’ll find the concepts intuitive.
Reading this guide in its entirety prior to doing your own analysis will be beneficial. After you’ve read it once, go back through and do the calculations step-by-step. That way, you’ll have a full understanding of how each component dovetails as you crunch your own numbers.
Remember that these numbers are unique to each series—which means you need to calculate them for each series you plan on advertising.
Now on to those principles.
Seven Fundamental Principles
Following these seven principles is critical to running profitable advertising, and can save you substantial time and money. As with all rules, there are exceptions, but if you’re starting off, you should break them only after careful consideration.
PRINCIPLE 1: Most, if not all, of your profit will come from sellthrough. This means the longer your series, the more money you can generally spend advertising it. Advertising a standalone novel is difficult (and rarely profitable outside of a BookBub Featured Deal), and not recommended unless you’re a highly skilled marketer.
PRINCIPLE 2: Focus your advertising on Book 1. Even in a series of standalone novels, many new readers will start with Book 1. Advertising later books typically produces smaller returns. Only advertise later books if you get a BookBub Featured Deal, or if you’ve exhausted all options for Book 1.
You can also advertise a series starter box set—e.g. Books 1, 2 & 3. The general principle here is that it’s easier to sell new readers on the beginning of the series (otherwise known as your funnel starter).
Advertising only Book 1 has the added benefit of making the administrative side far more manageable. Tracking ads for five or ten titles in a series is a nightmare.
As a final note, you should also advertise your latest release in a series—especially during the launch— but it’s much harder to analyze the data in the manner outlined below. There are too many variables skewing the launch numbers to make such analysis useful.
Instead, the general purpose of advertising the latest book is visibility and general awareness: making sure your fans are aware that it’s out, while also drawing some new eyeballs to the series.
PRINCIPLE 3: When using promotional sites, stick with proven options. You can find my curated, regularly updated list of recommended sites here.
PRINCIPLE 4: When running PPC ads, start small ($5 – $10/day). Make sure your ads are profitable before attempting to scale. If your ads are unprofitable at a small spend, a larger budget only increases your losses.
PRINCIPLE 5: While extremely useful, these metrics—cost per download, sellthrough, revenue per download, conversion et al—are estimates. Due to the nature of Amazon’s reporting and plain old variance, it’s impossible to predict with absolute certainty that your sellthrough will be precisely 41.4% over the next three weeks, that your book will convert at 32.5% on Tuesday, or a certain promo site produced exactly 921 downloads.
With that in mind, always factor in a margin of safety when advertising.
PRINCIPLE 6: Advertising most books will not be profitable (outside of BookBub Featured Deals). This is not a reflection on the book’s quality. Many books simply don’t hit the right commercial notes to resonate with a larger audience.
If you find, after running the numbers, that you’ve written such a book or series, don’t worry—but don’t force the issue by advertising anyway. This is how you rack up huge losses.
There are free marketing alternatives for these titles: mentioning them in your newsletter and autoresponder, entering cross promotions, social media, your website, and so forth.
PRINCIPLE 7: Profitable advertising requires consistent testing. You can—and should—use advertising principles, tactics, and ideas that have proven profitable for other authors. But ultimately, you must develop your own ads and ideas by finding what works for your books.
This is the number one rule of advertising. If you forget everything else, be sure to test.
One key note on testing: when adjusting your book’s back matter, blurb, cover, price, or other elements that affect the numbers, only change one major element at a time. If something impacts your numbers, you want to know exactly what was responsible.
That wraps up our key principles. Time to dive into cost per download.
Cost Per Download And Why It Matters
Everybody has a price, as the old saying goes.
Advertisers have prices, too. But the sticker price—whether it’s $20 or $400—isn’t an accurate gauge of the true cost. BookBub is sometimes mistakenly referred to as the most expensive advertiser in the game. At first glance, this rings true—but digging into the numbers reveals a different story.
That’s because determining an advertiser’s true effectiveness requires calculating the cost per download (or sale).
The equation for cost per download is:
The cost of advertising is the price of the promotion.
The # of downloads is the number of downloads or sales the promotion produced.
The baseline is the average number of downloads or sales your book gets without promotion.
To calculate the baseline, you can either:
- Take the average # of downloads over a recent 7-day span with no promotion.
- Look at the # of downloads a few days prior to the promotion and eyeball it.
Note that the baseline varies. Closer to release, or a couple months after a big promotion, your baseline might be 70 downloads/day. For an older backlist title, it might be 5 downloads—or even zero.
Let’s analyze some real numbers. Throughout this guide, I’ll be using my urban fantasy book Lightning Blade as a case study. Lightning Blade was a permafree title. However, it’s critical to understand that you can use the exact same calculations for paid titles without any changes.
Since I primarily advertise on Amazon, I’ve included Amazon downloads only. For the sake of the cleanest comparison, I do not include “tail” downloads from the days after the promotion. Depending on your situation, you might include downloads from other retailers—Barnes & Noble, iBooks, Kobo, Google Play et al.—or the tail.
Whatever you decide, make sure your calculations are consistent. Numbers that include all retailers and the tail are not directly comparable to those which only include a single retailer on a single day.
- Cost of advertising: The BookBub promotion cost $219.
- # of downloads: 19,266 Amazon downloads.
- Baseline downloads: about 300/day.
Thus, the cost per download was $219 / (19266 – 300) = 1.2¢ or $0.012.
By comparison, an EReaderIQ promotion produced a cost per download of $15 / (123 – 23) = 15¢ ($0.15).
Cost per download doesn’t just work for promo sites; in fact, it’s more critical when using PPC, where you can spend money like water.
The Facebook Ads produced a cost per download of $5.08 / (22 – 5) = 29.9¢ ($0.299).
Out of these three advertisers, BookBub was the cheapest, and Facebook was the most expensive. Knowing the true cost is critical when allocating your hard-earned advertising dollars. At the end of this guide, you’ll find a cost per download comparison chart of over twenty promotional sites.
Cost per download has far more utility than just determining the actual cost of advertising options, however. In conjunction with sellthrough and revenue per download, you can use cost per download to determine whether a specific advertiser will be profitable for your series—before booking a single promotion.
Before we go deeper into the math, however, let’s take a quick moment to address a question you might have: can I decrease my cost per download?
Yes—and I’ll show you how.
Sidebar: Increasing Your Conversion
Let’s say you’re getting 75 downloads for $10 (making your cost per download 13.3¢).
Not bad. Maybe you’re even profitable if your series has strong sellthrough. But you’d obviously prefer to get 100 downloads for the same price (10¢ a download).
This may or may not be possible when advertising on promotional sites. The biggest influences on the number of downloads you receive from these sites are the quality and size of their lists.
When you’re running PPC ads, however, conversion optimization has a huge influence on your cost per download.
Conversion rate is calculated as follows:
The # of downloads is the number of downloads or sales your book received.
The # of ad clicks is the number of people who clicked on your PPC ads. This tells you the number of visitors your Amazon page received.
To calculate (and optimize) conversion, you’ll need to run PPC ads like those offered by Facebook, Amazon, or BookBub. Otherwise, there’s no way to estimate how many people visited your Amazon page, since promo sites don’t provide this data.
The more downloads and clicks you get, the more confident you can be that your data is accurate. But this can take a long time and be very expensive.
My rule of thumb is to take a period of time with at least 1000 ad clicks and 300 downloads on a free book. On a paid book, the # of sales you’ll be dealing with is much smaller. It’s usually expensive to get 300 sales, so I’d aim for 30 – 100 sales.
These are Lightning Blade’s numbers from 6/24 – 7/15:
- # of downloads: 463
- # of Facebook Ad clicks (e.g. visitors): 1779
- Conversion rate: 463/1779 = 26%
Once you have your conversion rate, you can run tests to improve it. The following three elements have the largest impact on conversion (in order from biggest to smallest):
- Price. Lower prices produce higher conversion; free books have about a 10x higher conversion rate than paid.
Of these, you’ll mostly experiment with the price and blurb, since those are quick (and free) to change. By continually running tests, you can iterate toward a better Amazon page. Only change one thing at a time, so you know what was responsible for any changes in conversion rate.
An important side note: the audience to which you target your ads makes a big difference in conversion. For example, if I advertise my urban fantasy book to readers interested in thrillers, my conversion will be atrocious.
Make sure you’re only targeting audiences who will be interested in your book.
Now that you know how to increase your book’s conversion rate (and thereby decrease your cost per download), we can move on to the next piece of the analysis puzzle: sellthrough.
Sellthrough—also known as readthrough—is the lifeblood of every author’s career. Simply put, we need readers to read more than one of our books. Even if you never plan to spend a dollar on advertising, knowing your sellthrough is vital.
Sellthrough is the percentage of readers who, upon finishing a book, purchase the next book. Let’s say ten people bought Book 1, and five go on to purchase Book 2.
The sellthrough would be 50%—solid, but not earthshattering. Low sellthrough, however, is a glaring problem that you need to address ASAP.
Low sellthrough is generally a result of three things:
- Your book is free. Free books have significantly lower sellthrough (2%+) than paid ones (40%+). This is not a problem unless your sellthrough is low for a free book.
- Poorly optimized back matter. This can be fixed.
- People aren’t finishing the book, or didn’t find it compelling enough to pick up the sequel. This cannot be reliably fixed on that series.
Problem three often hints at deeper craft-related problems—or it could be that a particular series simply didn’t resonate with readers. If your sellthrough is consistently poor across multiple series, however, there are craft issues that must be addressed.
It’s important to understand that while sellthrough is fixable on a career level—you can learn techniques like cliffhangers, effective pacing, scene and sequel structure, plot structure and so forth to keep readers turning the pages—on a book/series level it usually is not.
I mention this because, in the face of poor sellthrough, the natural inclination is to rewrite or otherwise overhaul Book 1. This is not recommended. It’s a huge investment of capital, and rarely fixes whatever problems prevented readers from finishing the title.
Your best option for a series with low sellthrough is to conclude it and start a new series.
With those points out of the way, here’s the equation for sellthrough:
Let’s analyze some real numbers again from Lightning Blade:
- # of downloads of Book 1: 1405
- # of sales of Book 2: 20
- # of sales of Book 3: 11
- # of sales of the complete trilogy box set: 28
In this case, since the reader can either purchase Book 2 or the complete trilogy after reading Book 1, we need to add these sellthrough numbers together to get the total sellthrough.
1.42% + 1.99% = 3.41% sellthrough
This means, of the 1405 readers who downloaded Lightning Blade, 3.41% of them purchase the next book. That’s solid.
In a few minutes, we’ll use that sellthrough number to calculate revenue per download. But first, I’ll show you how to optimize your back matter to increase your sellthrough.
Optimizing Your Back Matter
While revising a book to increase its sellthrough is a dicey proposition that I do not recommend, optimizing the back matter is a quick, effective way to bump flagging sellthrough.
Most back matter is ineffective because it offers too many choices. Authors tend to overwhelm readers by throwing every link and request imaginable into the back matter.
This causes two problems:
- Each choice is not equal in value. A Twitter or Instagram follow—or even an Amazon review—has far less value than a sale.
- Having too many choices makes many people default to the easiest choice of all: closing the book without taking any action.
Thus, we want to streamline our back matter to focus on the essentials.
This is why I went from back matter containing three CTAs (calls to action)—a review request, mailing list link, and link to the next book (with a short blurb)—to a single CTA.
A link to the next book (with a short blurb).
The old, three CTA version is on the left. The new, one CTA version is on the right.
The new one was 1.4x more effective—producing 4.8% sellthrough (versus 3.4%).
This adjustment takes five minutes: just delete extraneous links and text. However, you’ve likely heard that reviews and mailing list subscribers are precious, and might be hesitant to remove such requests.
A review request has little value now that an automated Amazon pop-up at the end of each book asks readers to review it.
A subscriber has significant value, but a sale is often worth more. After a reader completes the mailing list sign up process some ten minutes later their interest in purchasing Book 2 might have already waned—or they might simply get distracted by other things.
I’d rather capture the sale right away.
I have a mailing list link in the front matter, for those readers who are interested in joining. I also include the mailing list link in the back matter of the final book in the series—if a reader makes it through an entire series, they’re likely to be a fan of my work and interested in hearing more from me.
You can use a two CTA version instead—a link to the next book (with a short blurb) and a link to the mailing list—if you’d like.
Finally, you can place any additional links (e.g. review requests, social media et al.) that don’t make the cut in the About the Author section.
Two best practices that are visible if you really squint at the image above:
- Put your CTAs right after THE END, on the same page. If you have a page break, an automatic Amazon pop-up selling other authors’ books and asking the reader to review triggers, essentially overriding your back matter.
- Always include direct links. This can be a pain if you’re wide, but including a direct link to the retailer (or your mailing list/the review form, if you include those links) helps.
Now that we’ve optimized our back matter, we can move on to our final bit of number crunching: calculating revenue per download.
Revenue Per Download
Remember that sellthrough number we came up with a few minutes ago? Good. We’ll need it to calculate revenue per download. This number tells you what each download of the advertised book (usually Book 1 or the series starter box set) is worth.
Here’s how you calculate revenue per download:
The ellipses indicate that if you have more books in the series, you need to add them to the calculation.
Here’s how this looks with real numbers—I’ve calculated the revenue per download at both 3.4% and 4.8% sellthrough so you can clearly see the impact.
- Book 1 net royalty: $0.00 (since it’s free)
- Book 2 net royalty: $4.99 * 0.70 = $3.49
- Book 3 net royalty: $4.99 * 0.70 = $3.49
- Box set net royalty: $9.99 x 0.70 = $6.99
- Book 1 > 2 sellthrough: 1.42%
- Book 1 > 3 sellthrough: 0.78%
- Book 1 > Box sellthrough: 1.99%
- Book 1 > 2 sellthrough: 2.54%
- Book 1 > 3 sellthrough: 2.15%
- Book 1 > Box sellthrough: 2.3%
At 3.4% sellthrough, each Book 1 download produces 7.7¢ in revenue if a reader buys Book 2, or 13.9¢ in revenue if they buy the box set. This averages out to 10.8¢ in revenue per download.
At 4.8% sellthrough, these jump to 16.4¢ and 16.1¢, respectively, for an average of 16.2¢.
One More Step If You Have A Box Set
While averaging the revenue per download numbers from the box set and Book 2 together yields a solid estimate, it’s not 100% accurate. This is because not everyone who buys Book 2 goes on to buy Book 3, which means this path through the funnel doesn’t produce the same revenue as when they buy the box set.
Calculating the accurate revenue per download number requires a weighted average.
Here, 41.6% of people purchased Book 2 (20 sales/48 sales of Book 2 + Box), and 59.4% of people bought the box.
Thus, to calculate the actual revenue per download you would do the following:
(0.416 * 0.077) + (0.594 * 0.139) = $0.115, or 11.5¢ per download.
The numbers don’t change much with the new sellthrough, since the # of units sold are closer to 50-50 (32 of Book 2, 29 of the box):
(0.524 * 0.164) + (0.476 * 0.161) = $0.162 or 16.2¢ cents per download
You can skip this step if you don’t have a box set.
Or just average the numbers together for the sake of simplicity.
That does it for the data analysis portion of the guide. Now that we have our numbers, we can determine what sites and platforms will be profitable for our books.
How To Analyze Profitability (Final Step)
Determining whether an advertiser will be profitable for a certain book is easy.
- Take that advertiser’s cost per download.
- Then compare it to your book’s revenue per download.
- If the cost is higher than the revenue, then that advertiser will be unprofitable.
For Lightning Blade, each download is worth 11.5 – 16.2¢ (depending on sellthrough). This makes any site costing more than 11.5 cents per download risky; recall from the introduction that these numbers, while incredibly useful, are just estimates. I have a lot of historical data suggesting I can expect 3 – 3.5% sellthrough on Lightning Blade. While I believe the improved back matter increased my sellthrough, I don’t know whether it’s actually 4.8%, or something smaller, like 4% or 4.5%.
I do know, however, that any advertiser costing more than 16.2¢ per download will be unprofitable.
Of course, calculating the cost per download for each individual site is a pain—not to mention expensive. It’s nice to just know what works, minus the trial and error.
So I did it for you.
Below, you’ll find 20+ promotional sites sorted by cost per download. Cost per download of course varies based on genre and your book’s conversion rate, but this should provide a solid starting point for your own endeavors.
These ran during the Summer/Fall of 2018 (so over the past four months) on Lightning Blade. The book had run on many of the sites before. If it was the first run, I made a note. I booked each site’s fantasy list (or the equivalent).
I’ve included numbers for a few other genres/promo stacks for the sake of comparison, as well as Lightning Blade’s 2017 BookBub run and a Freebooksy on one of my other urban fantasy books.
PPC platforms like Facebook, BookBub, and Amazon Advertising are not included, because cost-per-click (CPC)—and thus, the cost per download—is heavily influenced by your genre and the quality of your ads. When running PPC ads, you’ll need to use your own data to analyze your ads’ profitability.
The Process & Key Takeaways
Since we’ve covered a lot of ground—and this whole calculation business might seem overly complex—here’s the complete process, which also serves as a little to do checklist:
- Calculate the cost per download.
- Calculate the sellthrough for each series.
- Use the sellthrough to calculate the revenue per download.
- Compare the revenue per download to the cost per download. If the cost is higher, that advertiser will not be profitable for that series.
Run the numbers for a few series—I promise it’ll be an eye-opening exercise.
Other things to do:
- Focus your advertising on Book 1 (or a series starter box set of Books 1, 2 & 3).
- Test various blurbs, prices, and covers to increase your conversion rate (and thus lower your cost per download).
- Optimize your back matter to increase your sellthrough. Limit your CTAs—a link to Book 2 with a short teaser and maybe a link to your newsletter. Good sellthrough is the key to profitable advertising, and, in general, a successful career.
Remember to make just one major change at a time, so you can measure the effects.
If you’re having problems with sellthrough, but your back matter is optimized, read through the reviews and try to uncover why readers aren’t buying the next book. Don’t revise that book; wrap up the series, then start a completely new series that addresses the previous craft problems.
And finally, don’t force the issue. Many books can’t be profitably advertised (outside of BookBub Featured Deals). When this is the case, don’t spend money advertising them.
Instead, market these titles through free, passive means: your email newsletter and its autoresponder, social media, your website and so forth. This won’t generate explosive results, but it can produce a steady income, particularly with a large backlist.
You can find a curated, up-to-date list of promo sites (and their respective cost per download) on my site here.
That’s it. Go sell some books.
Appendix: Target Metrics
Numbers vary based on genre; romance, in particular, has its own set of marketing rules.
Remember: your own data is king. Nonetheless, if your numbers are well below the “solid” benchmarks below, you have a problem that needs to be addressed.
- Conversion (free book): 25%+ is solid; maxes out at 50 – 60%
- Conversion (paid book): 2 – 5%+ is solid; maxes out at 10 – 12%
- Sellthrough (from a free book to Book 2): 2%+ is solid; maxes out at 10% – 15%
- Sellthrough (from a paid book to Book 2): 50%+ is solid; maxes out at 80 – 90%
- Sellthrough (from a KU book to Book 2—note that I’m referring to readers borrowing Book 2, not buying it): 75%+ is solid; maxes out at 85 – 90%
Also note that sellthrough on later volumes in a series that must be read in order can approach 100%, unless you do something egregious like kill a main character.
Nicholas Erik is the author of 24 SF/F novels. He also runs marketing campaigns for indie authors in the mystery, adventure, and thriller genres. Check out his free Ultimate Guide to Book Marketing at nicholaserik.com/start.
A huge thank you from me to Nick for this wonderful resource – I’m sure a lot of work went into it and lots and lots of you will benefit from it. If you want more good stuff like this be sure to check out Nick’s site above, particularly the guide to promo sites he mentions in the post above. It’s another incredibly valuable resource – a handy list of worthwhile ad venues, tiered by effectiveness, and regularly updated. Be sure to check it out here.
With the sh*t hitting the fan over at Kboards since the new owners took over, it’s more important than ever that we authors share our methods and help each other whatever way we can, and ignore those who want to pull the ladder up behind them. Kudos to Nick for sharing this analysis and Happy Advertising to all of you!