Building an ebook business around analytics

What happens when a web publishing business enters the ebook space?

AskMen, “the leading online magazine for men,” has just launched an ebook publishing program, using the PressBooks Publisher Platform to manage the front-end catalog/website, and back-end ebook production. In the year-and-a-bit since PressBooks launched publicly, we’ve worked with many traditional book publishers, big and small. But what’s most interesting to us is non-traditional book publishers entering the ebook space, because they have the flexibility to approach book publishing in whole new ways.

Especially interesting to us are successful web publishers, mainly because web publishers have the most direct understanding of their readers, and reader behaviour. This skill, and approach, will be critical, we believe, as book publishing evolves. Web publishing is an analytics-driven business. In the ebook world, timely analytics are very hard to come by. And generally, analytics is not something most book publishers prioritize in their business. We believe this will change, as book publishing becomes increasingly digital.

In the interview below, I explore the issue of web vs. ebook analytics with Emma McKay, managing editor of AskMen’s online magazine, and the leader of their ebook publishing program.

Hugh McGuire: Can you define “analytics” for me, as it relates to your web publishing business? 

Emma McKay: We use a variety of metrics to assess the success of any piece of content we publish. Different sets of data, from Google Analytics to comScore data to various kinds of social engagement, shed different kinds of light on how readers discover and consume our content. We keep an eye on this information, and if an article’s not performing for us, we’ll try changing the headline or image or even the position of the piece on the homepage before we scrap it altogether.

HM: Why is analytics important for web publishing? 

EM: It reveals whether or not our efforts are effective. Analytics provides real-time insights that support our decisions, helping us ensure we are putting our energies into the projects that will deliver the greatest results.

Also, since AskMen is free to read, ad revenue is key for us, and advertisers need to know how their ads are performing. When we’re able to deliver real and impressive numbers, more interesting things become possible.

HM: How does analytics data from past web articles inform your decisions about the articles you publish now, and what you publish in the future? 

EM: Analytics data are not the only factor in our decisions, but they are a dominant force. AskMen has been around for about 13 years now, so we have some really old articles in the archives that rank well and continue to get a ton of traffic each month. We keep an eye on the data, and periodically update the articles that get the most traffic in order to keep them current. It’s easy to prioritize in terms of what’s getting the most views. If we can find new ways to build on those topics and give readers more of what they’re looking for (say, in e-book form), that’s what we’ll do.

HM: Other than the content, can you comment on how analytics influences how you publish on the web? Design? Article structure? Article keywords, tags, metadata? Etc? 

EM: SEO is a huge priority, and we work to keep abreast of changes in Google’s algorithms in order to maintain our rankings there. A couple of years ago, this meant we focused on high keyword density in articles, and now, it’s thankfully a little more subtle, which has allowed us to really improve the quality of the writing on the site.

We’re currently in the midst of a redesign process that takes analytics into account – user experience is key, and we want to give readers the best experience possible so they’ll stick around a little longer and return to the site on a regular basis. If we position related articles (which appear as “related” because of the tags they are given) to the side of the main article text, will people be more likely to click on them than if they were to appear at the bottom of the page? Analytics can give us the answer. Analytics help us decide whether a 1,000-word article should appear on a single page, or be spread across two or three pages, because it gives us insight into how readers are consuming content that we can apply immediately.

HM:  What are the key metrics that you measure to define a successful web article or set of articles?  

EM: We mainly look at unique visitors, page views, CTRs, Facebook likes and tweets. Bounce rates can be very revealing, too. We also pay attention to where traffic is coming from so we can better understand how we should promote our content. We watch for comments on articles as an indication of reader engagement.

HM: Turning to ebooks — can you tell me about your experience with the analytics data available to you as an ebook publisher? 

EM: We were hoping to get access to real-time sales data, but so far this has proven impossible. We do have access to almost-real-time data on sales trends, but it does not give us a completely accurate picture of where we’re at. We’re accustomed to being able to immediately track the impact any action has had, so not being able to tell whether promoting something on the homepage or in a newsletter has had an immediate impact on sales is very frustrating.

HM: What sorts of ebook data would you most want access to? 

EM: Real-time sales data is the top priority by far, but we’d also love to know how readers discovered the ebooks, and whether they purchased a single ebook or more than one. The big retailers are all collecting data on who is buying ebooks and how ebooks are being consumed, but they’re not about to share this information with us. So we’re also pursuing direct sales opportunities – among other advantages, this allows us to get to know our readers a little better.

HM: How would that data influence your ebook publishing decisions? 

EM: The better we understand our readers, the better we can cater to them. Who are they, and where do they live? It’s always fascinating to track how content we publish on the site gets picked up in different corners of the world. Perhaps our ebook audience is entirely based in India. Or the Midwest. Either way, we’d like to know, so we could tailor our program accordingly.

We’d like to know whether our audience consumes ebooks in the same ways they do our website content – are they dipping in and out of them to get the advice they’re looking for, or are they sitting down for a couple of hours to read a narrative like Pakistan Chronicles from beginning to end? Are they looking to go deeper into a single subject, like they can with our titles Understanding F1 and Build the Ultimate Watch Collection, or are they looking for the kind of guidance we can offer with The Hair Manual (coming out this week) or a how-to-make-fitness-part-of-your-daily-life-for-life title like our recent hit Mission: Motivation?

We’re also interested to learn how readers discover our ebooks. If we could see that they were flocking to the ebooks from a specific source, we’d take a closer look at that source, and at ways we could build on whatever it is that’s working, whether this leads us to promote the books in a specific way or through a particular outlet, build new partnerships or establish a themed series or imprint.

HM: Given the paucity of timely ebook data, do you think that publishers should still be thinking about analytics as they build ebook publishing programs? How? 

Absolutely — it’s essential to keep close tabs on analytics, especially in relation to promotional efforts, so we can identify what works and what doesn’t.

We can make timely publishing decisions with ebooks, but we’re not going to chase fleeting trends. Analytics can help us narrow in on the trends worth engaging with. We need to know what people are looking for, and from this information we can begin to figure out how to deliver a product that our readers want but haven’t even realized they’re looking for yet.

tags: , , , , , , , , ,