Welcome new readers! If you would like to subscribe to the Bard Press Essays, you can do that here.
Bard Press Essays #10
April 14, 2021
1.
Launching a book is difficult and book launches are getting more difficult.
For as long as I have worked in publishing, I have wondered what makes some books work and others don’t. I am sure everyone who works publishing books wonders the same things. When things shift, like they did in 2020, you wonder a little more—what is still true about your incomplete view of how it all works.
Book launches have been hard to do in the last 12 months. As the reality of the coronavirus settled in last March, publishers made decisions to push many releases later into 2020 or fully into 2021. Those pushes made sense to me with many venues either unavailable or overwhelmed for much of last year. Physical bookstores were closed. Media outlets were fully committed to COVID and the US elections. And readers weren’t reading as many books. They were searching websites, twitter feeds, and podcasts for reporting on the current events.
That doesn’t mean books weren’t being purchased. The business category was down 11% for the year, but that still meant that millions of copies were purchased last year. With less launches though that meant that books that were being purchased were already published. That might sound obvious but book publishers carefully measure the success of recent launches versus their full catalog of published titles. If you are reading this, you’ve probably heard the terms frontlist for titles that are less than a year old and backlist for titles that are more than a year old.
Historically in book publishing, the backlist had more sales volume than the frontlist titles, but not by much. The trend in the last ten years has seen the backlist growing as a percentage of total sales. In the business category, backlist sales have always been a larger percentage of sales. You can see this in the appearances of popular titles on the bestsellers lists week after week. As you might imagine, this backlist growth trend accelerated with fewer launches in 2020. In business, 86% of sales were backlist titles. When you look at the dozen titles that sold over 100,000 copies in the business category last year, none were published in 2020.
2.
It’s hard to define success when you launch a book.
A book launch is a set of activities to engage the public and create momentum. I have been involved with the launches of hundreds of books and there is no common blueprint for success. Each book is different—in its approach to a problem and delivery of solution. Each author is different—in what they bring to the launch. And the world itself is different every time you bring a book into the world. Using my “every book is a startup” metaphor, each new business is going to take a unique approach in delivering value to customers.
I feel like what we miss when we launch a book is a hard goal to move towards. In startups, founders are searching for the best way to grow and often trying many different things. Their goal early on is often growth itself, but profitability, market share and long term plans can all influence their decisions. With books, we seem to leave it open ended. Good launch plans certainly assess all of the assets available, but it is a forecast that is adjusted as the book moves through the launch cycle.
The chief constraint is time. Publishers will devote effort to a book for a certain period of time in an attempt to generate that momentum needed. Authors many times follow their publisher’s lead, devoting most of their effort in the lead up to and follow-on after the launch. The best authors will leverage all of their connections and do everything they can to generate demand and give their book its best chance at the launch. The time spent is measured in months and in that short time, the book needs to find its audience.
3.
Let’s find a helpful goal.
My interest in defining a different and definitive goal for book launches has many reasons. Goals are highly effective at generating action when the exact path forward is unknown. Authors are only going to write a handful of books, so giving them an informed goal lets them move toward success without needing the experience of a dozen launches. Authors can also more accurately measure their ongoing success and evaluate the various returns on activities to further promote the book. For myself, the line of research helps me think about the books that will work best in the Bard Press model of one book each year.
So, I started with the question of whether the sales of past titles can tell us something about forecasting future success. Looking at sales of titles is difficult because of the nature of publishing that generates a few big, big hits and many, many small sellers. That enormous range in sales makes it very difficult to make comparisons using normal statistical techniques like means or standard deviation.
In the last few months, I took a new approach to the question. I decided to gather data on a set of titles and look at two things. First, I want to know how many copies the titles sold in the first year it was in the marketplace. I created a set of cohorts based on the first year sales:
- Titles with less than 10,000 copies
- Titles with 10,000 to 25,000 copies
- Titles with 25,000 to 50,000 copies
- Titles with 50,000 to 100,000 copies
The second piece of data I collected for each title was the total lifetime sales and I divided up the results into similar cohorts:
- Titles with less than 10,000 copies
- Titles with 10,000 to 25,000 copies
- Titles with 25,000 to 50,000 copies
- Titles with 50,000 to 75,000 copies
- Titles with 75,000 to 100,000 copies
- Titles with 100,000 to 200,000 copies
- Titles with more than 200,000 copies
So, think of first year sales as an input variable (X) and the lifetime sales (LTS) as an output variable (Y). Rather than any fancy statistics, I just wanted to know what bucket each title fell into and think about the probability of each of these conditions.
My dataset was a group of 6,775 titles in business and self-help, published between 2012 and 2015. These years were relatively stable for those categories and this range allows each title to have at least five years of lifetime sales data.
| First Year Sales | Number of Titles | >10K Lifetime Sales | >25K Lifetime Sales | >50K Lifetime Sales | >100K Lifetime Sales |
| <10K | 6037 | – | – | – | – |
| 10K-25K | 504 | – | – | – | – |
| 25K-50K | 140 | – | – | – | – |
| 50K-100K | 94 | – | – | – | – |
4.
The magic number
Every industry has its rules of thumb. Book publishing is no different.
One number that comes up often is 10,000 copies. Many people have attested to the importance of reaching this five figure sales point for any book. Often, the reasoning was framed around the idea that if you could get 10,000 copies into the marketplace, you’d truly know whether a title could find momentum. Others have told me that 10,000 copies is where momentum takes over and the book starts to sell itself.
So, I chose 10,000 copies as the lowest threshold for the cohorts. It made sense as a starting point to use that institutional wisdom.
The first slice of data partially confirms book publishing’s rules of thumb: if a book doesn’t sell 10,000 copies in its first year, there is only 11% chance that title will ever sell more than 10,000 copies. There is only a 2% chance sales will exceed 25,000 copies and it gets worse from there.
| First Year Sales | Number of Titles | >10K Lifetime Sales | >25K Lifetime Sales | >50K Lifetime Sales | >100K Lifetime Sales |
| <10K | 6037 | 11% | 2% | 0.3% | 0.1% |
| 10K-25K | 504 | – | – | – | – |
| 25K-50K | 140 | – | – | – | – |
| 50K-100K | 94 | – | – | – | – |
Finding #1: You have to sell more than 10,000 copies in the first year, or it will be difficult to ever exceed that sales level.
5.
Is there a magic number?
Selling less than 10,000 copies is definitely a barrier to long term success.
The next question is whether selling more than 10,000 copies accomplishes anything further. Does it build momentum?
Take a look at the data below for the 10K to 25K first year sales cohort:
| First Year Sales | Number of Titles | >10K Lifetime Sales | >25K Lifetime Sales | >50K Lifetime Sales | >100K Lifetime Sales |
| <10K | 6037 | 11% | 2% | 0.3% | 0.1% |
| 10K-25K | 504 | N/A | 42% | 13% | 2.4% |
| 25K-50K | 140 | – | – | – | – |
| 50K-100K | 94 | – | – | – | – |
What the data says is that if you can get into the 10,000 to 25,000 copy range for first year sales, you have 42% chance of selling more than 25,000 copies in lifetime sales.
Sit with that for a minute.
If you get past that 10K mark, there is a 4 in 10 chance of getting beyond 25K copies sold.
So, all those people who told me 10,000 copies mattered were right.
Below that level, you get stuck. Above that amount, you create a significant opportunity for continued success.
If you were thinking about investments in both time and money for a book launch and you look at this data, you see the incredible importance of early efforts to reach the 10K threshold and the strong probability to have continued success with your book.
And this trend doesn’t stop there.
We see the same thing in each of the first year cohorts after you pass the 10,000 copy market:
| First Year Sales | Number of Titles | >10K Lifetime Sales | >25K Lifetime Sales | >50K Lifetime Sales | >100K Lifetime Sales |
| <10K | 6037 | 11% | 2% | 0.3% | 0.1% |
| 10K-25K | 504 | N/A | 42% | 13% | 2.4% |
| 25K-50K | 140 | N/A | N/A | 47% | 19% |
| 50K-100K | 94 | N/A | N/A | N/A | 50% |
If you can break 25,000 copies, there is a 47% that you can get passed 50,000 copies. And if you can sell 50,000, there is a 50% if you could get achieve 100,000 copies in lifetime sales. Again those are very favorable odds for the author considering further support and investment in the promotion of their book.
Finding #2: If you sell more than 10,000 copies in your first year, you have a 40% – 50% probability of selling enough copies to moving into the next lifetime sales cohort.
6.
Timeframe Matters
The timeframes I chose for this analysis are important. For the input variable, I choose first year sales. I was interested in looking at a longer window of time to assess the early success of a book. This matches the traditional measurement of a frontlist in measure of a title’s sales in the first year. As publisher, I want to be able to tell the authors I work with that there is a strong payoff to support 12 months of effort after the book launches.
One problem from watching so many book launches over the years is too many books hit the market with a big launch and fizzle quickly. The author lines ups sales with individuals and organizations in the opening weeks, often to make a bestseller run. In the following weeks, there is few follow-on sales or continued conversation in the marketplace about the book. I have come to think that authors can be too effective and efficient at pulling all of the demand out of the market so early in the life of the book.
That leads to a great follow-on question – Do launches matter?
I’ve started to do research to answer that question.
As a publisher, I firmly believe launches matter. Books have the unique ability to draw attention to you, your company and your idea and draw it into a specific window of time with a book launch. This makes publicity effective. Connecting your launch to a time of year (i.e. January – New Year, New You; February – Black History Month) can build on interest that already exist in the marketplace. Launch dates focus customer actions.
My preliminary analysis on launches sales, as I define as the first ten weeks after publication, shows those early sales are less predictive of lifetime sales. Books that sell 10K to 25K copies in the first ten weeks and then go onto sell over 25K copies only 22% of the time, that is about half as often as a book that sells the same number of copies over 52 weeks.
This first analysis would indicate that a book that sells consistently (i.e. 300 copies a week for 50 weeks) is better positioned for long-term success versus another book that sells most of its 15,000 copies in the opening 10 weeks. That makes intuitive sense to me. The launch of the book is about author and their ability to generate interest in a new project. A book that sells consistently taps into a different kind of demand—demand that might be driven by the author, consistent recommendations from other readers, or adoption in organizations and educational institutions.
On this question around launch sales, I need to pull a bigger dataset and do more analysis before I can speak with the same confidence as the relationship between first year sales and lifetime sales.
7.
The Goal
I started this research to better understand launches. I wanted more insight into how to overcome the growing dominance of the backlist. I also wanted more data to support authors and show the benefits of investing in long-term promotion of their book. The data is amazingly clear.
The goal is to sell 10,000 copies consistently in the first year the book is on sale.
How do you do that? That’s the topic of next week’s essay.
