Growth hacking as a book publisher

by Frank Proctor and Jason Li

We run Muse, an indie publisher based in Hong Kong that specializes in cultural nonfiction and translated literature. As a small international publisher, we have reaped big benefits from the disruption of traditional publishing. Originally rooted in print, we have now been publishing ebooks for four years – not long enough to be an expert by any means, but long enough to understand that digital media and the internet will play the central role in the way stories are told, and the way writers and readers interact.

Being on the front lines as a small-time publisher, we admit that we don’t have a roadmap for where the industry is going; so we have little choice but to experiment, learn and iterate – like an explorer feeling his way along the wall of a dark cave. Reading Peter Sims’ book Little Bets was instrumental in shaping this intuition into a business philosophy of placing “little bets” and learning from them.

We recently published Snow and Shadow, a collection of short stories by an award-winning Hong Kong Chinese writer, Dorothy Tse, translated into English by Nicky Harman. Launched in Hong Kong and the US/UK in March and June respectively, Snow and Shadow is the latest stage of our learning process.

This is the first time we have created a dedicated book website and Facebook page; the first time we’ve had a Tumblr presence; and the first time we’ve built an app around a literary work. Although Muse already had its own Facebook page and modest website (far from state-of-the-art), we chose to invest our digital resources mainly in the book, not in the publisher’s brand.

Our ultimate goal is, of course, to sell more copies of the book – both electronic and print. But beyond that, we have three important goals:

  1. Connect Dorothy Tse’s writing and ideas with a wider audience;
  2. Generate “buzz” around the title – reviews, engagement, shares;
  3. Learn! (Build expertise in digital and online publishing.)

We are fortunate to be working with a wonderful author, who was open to having her work taken into new creative directions; her clear prose style and vivid, surrealistic imagery offered many intriguing possibilities. We also had terrific publicists at DeBartlo & Co. who were very proactive and full of ideas.

What we learned from our little bets

A free iOS app attracted downloads but failed to “convert”

We turned one of the short stories in the book into an iPhone app

An attractive and free app (illustrated by renowned Hong Kong artist Chihoi) proved to be a great magnet for attention. Through a combination of social media marketing and Facebook ads, we have gotten 120 downloads of the iOS app (so far). From the outset, we decided to create the short story app as a standalone experience that would be enjoyable without any knowledge of the book, so we didn’t expect a big bump in book sales. But we were disappointed when there wasn’t any upswing in book sales. Aside from a mention in Time Out Hong Kong, the app did not manage to generate significant buzz in the press or engagement within online communities.

There are several factors that we believe contributed to the lack of conversion and engagement:

  1. The in-app link to the book is (intentionally) subtle;
  2. Because we decided to stay within the Apple ecosystem, the in-app link points people to iBooks, although that is not usually a high-sales platform for us. (To make matters worse, users without iBooks on their iOS devices are directed to install iBooks, rather than to buy the book);
  3. The app itself may not be intuitive to use; we originally created it as an interactive website and only later ported it into an iOS app, which has affected its usability on iOS;
  4. We didn’t define upfront what “engagement” we wanted from users. Making clear calls to action (e.g. share, comment, ask us anything) might have solved our engagement problem.

The elephant in the room when it comes to mobile apps is, of course, the budget. We started by building an interactive website because we feared we didn’t have the resources to build and maintain an iOS/Android app. Only later did we discover a beautiful loophole that allowed us to port the website into an iOS app: keep the scope modest and use Xcode (Apple’s development toolkit)’s native, drag-and-drop tools. In this way, we managed to launch the app without breaking the bank.

1:5 is not a bad ratio for review copies

With the help of DeBartlo & Co., we sent out plenty of review copies to both “professional” and “amateur” reviewers (though the line between them is increasingly blurry).

About 150 printed books and 10 ebooks went out to mainstream book reviewers and bloggers, and got us eight reviews and interviews so far:

We sent out 80 printed books and 40 ebooks to selected Goodreads/Amazon reviewers (after emailing them first to see if they were interested). So far, that has generated 7 reviews on Amazon and 16 ratings/reviews on Goodreads, as well as a few mentions on readers’ blogs.

To sum it all up, we got a 1 : 20 ratio of reviews to review copies that we sent out to mainstream media and literary blogs; whereas it was a 1 : 5.2 ratio for copies to Amazon/Goodreads reviewers. This was clearly a good investment of time and money (particularly because the reviews from both professionals and amateurs were overwhelmingly positive).

Why are we discussing review copies in an article on digital experimentation? That’s exactly the point – we failed to think digitally. We had a very traditional view of the process of sending out advance copies for review. Fortunately, DeBartlo & Co. helped us create PDFs which could be sent around with ease in response to requests for digital copies. But as the publisher, our own planning was too traditional.

Facebook ads work for app downloads, but that’s about it

For Snow and Shadow, we invested in online ads to promote the book website, the book’s Facebook page, the Kindle ebook, and the iOS app:

  • Google Adwords worked to get people to the book website. We used search ads with remarketing, rather than display network ads (which based on past experience can generate a flood of clicks from dubious sources), and we intentionally bid higher for competitive keywords. At an investment upwards of a thousand dollars, Adwords was responsible for about 70% of the traffic to the site during the first few months. But at over $5 CPC (cost-per-click), we did not get the ROI to justify continuing the campaign – only 23 website visitors clicked on our link to Amazon to make a purchase. Campaigns for two of our earlier books, where the ad links went directly to the Kindle page, had been only marginally more cost-effective. So after three tries, we have yet to achieve satisfactory ROI promoting a book with Adwords.
  • Facebook worked to get likes on a Facebook page, and to get attendance at our launch reading event. However, a Facebook ad campaign in conjunction with Kindle pricing tests (ranging from $1.99 to $9.99) generated plenty of clicks at $0.81 CPC, but with negligible impact on sales. Again, the ROI wasn’t there.
  • Facebook ads successfully led to mobile app downloads. We supported our app with a small-scale, one-month Facebook ad campaign, at a total cost of $330. iTunes recorded 70 downloads during that period, which meant the campaign cost us $4.70 per download. Contrast this to our $5 CPC for the Adwords campaign – an app download is a multi-step process, yet it ended up being cheaper than a single-step click. (Oddly, our Facebook ad analytics did not register any downloads, which may have been because of the way we set up the campaign, but the downloads were clearly resulting from our Facebook ads.)

Overall, with the exception of mobile app downloads, both Google and Facebook ads were disappointing in terms of fostering actual engagement (book sales and comments/likes on posts).

Twitter and Tumblr are nice, but nothing exciting happened there

We opted to use our own personal Twitter presences, rather than one for the company or book. At the time of release, both of us had Twitter accounts with hundreds of followers and neither of us was keen on company/non-person Twitter accounts (we don’t even follow other cool indie publishers!). Our publicist suggested trying a paid Twitter promotion, but we decided against it after noticing a backlash against them among our peers. There were around 75 organic book-related tweets/favorites/retweets, but not the kind of widespread sharing or engagement we had hoped for.

In contrast to our experience on Twitter, we only started our Tumblr presence at the launch of Snow and Shadow. Our Tumblr is branded “Muse,” and serves a dual purpose as a company blog and a place to engage with the larger Tumblr community. As the former, it’s a great tool to quote reviews of the book, put up screenshots of the app, and feature visual artists that we work with. As the latter, it’s been a struggle starting from zero followers, particularly because we lack the resources to create high quality, custom content for the book reading and writing community.

Writer & reader online social networks have potential

We posted individual short stories on Wattpad (45 reads and 2 votes) and Scribd (35 views, 1 like). These numbers themselves aren’t bad given our lack of previous activity on either platform. But again, it is hard to link this either to positive commentary about the book, or to engagement with our other online presences.

On the other hand, as we mentioned above, we were able to generate a good amount of buzz and activity on Goodreads by sending out review copies of the book, with some positive overflow to reader blogs and Amazon.

How our little bets will benefit our next book release


  • Dorothy Tse has gotten more international attention, including some invitations to international literary festivals, which is raising the profile of Hong Kong writers as a whole.

Jury still out:

  • We track our Kindle and iBooks numbers daily, and our ebook sales have fallen short of expectations. But there’s a significant time lag before we get a clear picture of print sales, since we are distributing internationally from Hong Kong. So we aren’t yet sure how our digital experiments may have affected print sales.

What we’ll do differently next time:

  • Spend less on online advertising. Several dollars a day in online advertising can add up, and campaigns can proliferate until you have spent much more than you had planned. By cutting down on online advertising, we’ll have more to spend more on digital content and presentation.
  • Repurpose Tumblr for the next book, and continue to build a following (even if we don’t do any more than post press releases and excerpts). If we can create a little more original content, that would help too.
  • Think more strategically about how we balance the publisher’s digital platforms versus individual books: when do standalone Facebook/Tumblr/websites make sense and when are they fragmenting and harmful?
  • Better integrate digital planning into our review copy program. For our next book release, leverage ebooks to get more copies out earlier, expand our mailing list, and make it more cost-effective by sending more ebooks and fewer printed ones. We can also utilize a service like NetGalley to post review copies online for download.
  • Use what we’ve learned to develop an enhanced app. We got a good education in how to represent text and narrative in a digital format. Next time, we can focus more on making the app easier to share, interact with, and give feedback on.
  • Continue our learning process with at least one “blue sky” project next time. For this, we’ll try to think digital much earlier, reading each story with a fresh pair of eyes, looking for a practical digital angle. Can digital add something new, something that was not possible with print, to a particular story?

Frank Proctor is publisher at Muse.

Jason Li is a Hong Kong-based designer.