Despite its speed and size, the giant wave of digital book discovery has not yet crashed the shore. Sure, we know it’s coming. Hell, we can see it coming. But that does not mean that as of today (or this season, for that matter) digital readers (both people and devices) will be overtaking our beaches with their referrals and likes and retweets.

Don’t get us wrong: we here at Melville House welcome the wave, and all it has to hold, as much as any publisher. Lord knows benefits exist when readers discover books just as often through social media and other online efforts as through IRL-word-of-mouth-neighborhood-bookstore recommendations. But it seems that day is not yet upon us, at least according to a recent study conducted by the Codex Group:

Two years ago, 35% of book purchases were made because readers found out about a book in bricks-and-mortar bookstores, the single-largest site of discovery. This year, that figure has dropped to 17%, a reflection both of the closing of Borders and the rise of e-readers. In the same period, personal recommendations grew the most, to 22% from 14%. Some three-quarters of personal recommendations are made in person, while the rest come by e-mail (8%), phone (7%), Facebook (4%) and other social networks (3%).

Even though it’s not news that more and more readers are becoming “hybrid” readers, meaning they read and learn about books both online and off, the Codex study shows us that despite the hubbub, digital discovery does not carry as much weight as personal recommendations. What’s more, the study concluded that people who read on tablets, versus dedicated e-readers, buy fewer books and spend less time actually reading, which speaks to the theory that tablets, while terrific for a mobile experience, bring with them a Web mentality — more frenetic browsing than title-focused reading.

Shelf Awareness has more:

Surprisingly, considering all the attention it’s gotten, digital mass media, including Facebook and Twitter, rose just to 4.5% from 1.9% as a place people learned about the books they have bought. And the online channel represents 9% of discovery, which Hildick-Smith called “way underperforming” in light of the amount of purchases made online. In part, this is because many readers search for books online knowing what they want. (By contrast, readers tend to go into bookstores with an “open mind.”) The result, Hildick-Smith said, is that many books “get lost in the long tail.” Amazon, for example, has 32 million book offerings.

More than 50 million people own e-readers in the United States alone. The United States also has the world’s largest publishing market, followed by China, Germany and Japan. As the map and chart below show, such numbers stir up a recipe for digital discovery more potent than anything we’ve seen before. It remains to be seen, though, if book purchases will correspond with the number of devices sold.

The question now, as ever, is how publishers can sculpt a model that takes advantage of the coming digital wave while at the same time supporting the lasting power of personal recommendations.

A final note from the Shelf Awareness report may contain a key insight: “A problem for publishers and authors of new titles is that the vast majority of personal recommendations are backlist titles. Only 6% of books recommended personally have been published in the past half year–and just 2% were published within three months.”