Corner Office: Merriam-Webster Adapts to a Mobile World

Reference publishing underwent a massive transformation when the internet hit the scene in the early '90s. Consumers began to view the web itself as a reference tool instead of a bound dictionary. In light of this change, and the emergence of sites like Dictionary.com, publishers had to adopt smart online strategies, relying on pageviews and display ads to pay the bills. "As early as 1996 we decided that was going to be our strategy," says John Morse, president and publisher of Merriam-Webster, and advisor to Book Business. "In the '90s we moved pretty aggressively into having free online access to our dictionaries." That effort has made Merriam-Webster.com a go-to resource for word lovers, and it now captures over 100 million page views each month.

Today, Morse notes, the reference industry is undergoing another transformation -- a shift from large desktop screens to small mobile displays. Reference users expect dictionary access on their smartphones and within their ereaders, which has spurred Merriam-Webster's meticulous development of dictionary apps and ebooks. Though the company has found success in the app space, with seven apps launched and more to come, ebooks have proved a tougher nut to crack.

"Reference in a lot of ways was an after thought in the ebook space," explains Matthew Dube, vice president of business development at Merriam-Webster, "I think the ebook was initially conceived as a 'start at page one, end at page 200.' In reference it is obviously a different animal than that."

Although Morse and Dube are pushing for EPUB standards that support reference, they are exploring other opportunities for Merriam-Webster. Morse sees a bright future for reference books that cater to English language learners, an international group that numbers in the billions. Likewise, creating more gamified apps and engaging web features provides an opportunity for Merriam-Webster to foster a lexically-curious community that it has not yet fully tapped.

Here Morse and Dube detail Merriam-Webster's latest digital transformation and the opportunities that have come with it.

What are some technical difficulties you've run into with ebooks?
Matthew Dube: EPUB does not allow for what we call "headword lookup." So if I'm reading a novel and search the word "ubiquitous," it's going to give me every instance of the word "ubiquitous" in that text. Often it is searching through a very large file to give every bit of context possible.

When someone is using a dictionary, and they look up the word "ubiquitous" what they want is a definition. They want to see more information about that word. EPUB does not at the current time allow for that type of user experience. If someone purchases an EPUB version of the dictionary, when they search "ubiquitous," they will get every instance of the word within the entire dictionary. That's not really a good experience for the user.

Where have you found success with ebooks?
MD: The real winner there is the Kindle platform. That is because their technology is something they purchased a while ago called Mobipocket. That does allow "headword lookup" and it allowed Kindle to incorporate that technology into their entire system for their E-Ink readers. If I purchased a new Kindle E-Ink reader and I wanted to purchase the Collegiate Dictionary, I would have that as a stand-alone title in my library but also have the ability to set that as my default lookup reference when reading a novel. We're trying to work with other platforms to get that type of feature built in.

John Morse: You can note the irony here. The reason that we had some success with Amazon and the Kindle reader was precisely because they did not embrace EPUB standards. And I think you're going to hear from a lot of other publishers that EPUB is very important to ebooks.

Why is EPUB less suited to reference books?
JM: I think in part it is because reference books and textbooks rely more heavily on the generically tagged databases underneath them. That's what I think the EPUB standard has trouble with -- trying to retrieve not just a word but a word that has been tagged in a particular way. Textbooks would have the same problem. I think right now EPUB is more heavily skewed toward running text like a novel versus text that is heavily tagged.

Have you had problems with discoverability in the app space?
MD: We haven't had that problem. We've been fortunate because we've been very deliberate about the user experience and about the product. Rather than racing to market with the apps we've been very considerate in developing them and have used a lot of consumer data from our website. Because of that we're able to cut through the noise.

JM: I also think we are working at an advantage because people expect there to be a dictionary app. I think people come looking for us simply out of the expectation that if dictionaries have been on every other digital delivery platform, surely they will be on this as well. Motivated users come looking for us. I think that gives us a slightly easier job in terms of discoverability.

Any new apps on the horizon?
JM: I'll let Matt talk about our newest app, but let me provide a little frame here. We've seen a number of different opportunities to pursue -- moving from being a domestic publisher to an international publisher to focusing mostly on native speakers of English to focusing on English language learners as well.

But another area that we've had some success in and want to build on is the recreational sense of language. There is a motto that I keep repeating: "Language is serious. Language is fun." Obviously for years and years we have been pretty much on the "language is serious side" of that continuum. A while ago we started introducing vocabulary quizzes onto the website and they were a very big success. Now we have carried that into the app world.

MD: That manifested itself into an app called Quizzitive, which we launched last fall. That's a stand-alone application that is designed to help folks learn vocabulary while at the same time having fun with words. It has been a great success so far and we're working on the second iteration of it now. We wanted to package those quizzes that we knew were really popular. The quizzes that John was alluding to on the website have a completion rate well over 90% so when someone starts a quiz, we know that they're finishing it. We added some adaptive learning to the application so that people are working through these words and getting to know them better.

It sounds like there is a lot of data mining behind both the site and your apps. Is data collection a big focus for Merriam-Webster?
JM: Yes. You've put your finger on what is in some ways the biggest game changer in dictionaries. Up until the 1990s when websites like our own started getting enough traffic for the statistics to have some meaning, dictionary makers didn't really know what words people looked up the most in the dictionary. Now, knowing that, we really have a much richer sense of why people use dictionaries, what kind of words they are interested in, and what kind of things are spurring dictionary use.

It has changed the way that we think about dictionaries. One of the most gratifying aspects of what the statistics really tell us -- what we hoped was true but weren't positive was true -- is that dictionary users are fairly serious people who are looking up words to answer fairly serious questions. That is exactly what the statistics bear out.

If you look at the four-month list of words on the site, you don't find a lot of slang words. You don't find a lot of technical words. You don't find a lot of newly coined words. Marketing departments for dictionaries always told us that's what people want. They want to know that you've updated it with all the most recent teenage slang or the most recent tech term. We do that. I have to tell you, that's the easy part of lexicography. The hard part is making sure that if you look up "oxymoron" that we in fact have detected the subtle shift in the way that word is being used. It was a delightful verification to find that which we worked so hard on was being most sought out by our users.

What's the rough revenue breakdown for your desktop, mobile sites, and apps?
JM: Right now the vast majority of revenue, including print, is coming from ads on the main Merriam-Webster desktop/laptop website. Probably next on the list is advertising on the mobile website where we have been very successful. The third on the list I would say is either advertising in the free apps or the revenue from paid apps. That's just a snapshot of where things are. I think we fully expect and are planning strategically for that picture to change. We are well into a transition from the dominance of the fully featured websites to more and more use of mobile devices, whether that's through an app or mobile website.

What are the biggest challenges Merriam-Webster faces today?
JM: I do think it is that transition [from desktop to mobile]. We were very successful with the transition from print into web-delivered dictionary content. It was a slow-motion transition that started in the '90s and has culminated today. Having done that, we have to go through another transition which will be just as challenging. That transition is moving from the big full-featured website being the legacy product and moving into smaller, more specific sites.

The experience now is much more likely to be paired down -- a basic exchange. You're more likely to come to a small screen rather than a big screen. How do we negotiate that in terms of user experience and in terms of a business model? The plain fact is I don't think we have all the answers today just like we didn't have all the answers to the print-to-web transition that took place 20 years ago. But because we have that basic connection with users and a clear sense of what the utility is and what is the real benefit of coming to the site, I don't think there is anything to fear there.