Deborah Cooke: Author

Once upon a time, we bought printed books in bricks-and-mortar bookstores. We went to the store, maybe with a particular title in mind, found the book and bought it. If you’re at all like me, a trip to the bookstore also meant buying other books, maybe books that weren’t on the shopping list.

This sounds simple, but actually, it’s quite a complicated activity – particularly if you need to replicate it for an online bookselling portal. Some parts of it are easier (searching by title, searching by author, searching by genre) but others are tough (browsing, recommendations of similar authors). The tougher ones tend to be the ones that lead to more impulse shopping, so they are of interest to anyone in the business of selling books.

Another term that is used for this process of browsing and finding books is discoverability. In those bygone days of shopping for print books in bricks-and-mortar bookstores, there were established methods for helping readers discover new authors or new titles. It turns out that a lot of those methods are based on the experience of moving physically through a bookstore, and don’t necessarily translate well to online shopping. When they do translate, they mutate in some interesting ways.

1. Specific Title or Author
It’s pretty easy for online bookstores to replicate the search for a specific book. This was the easiest shopping pattern to mimic and so it was the first to appear on online shopping portals. There’s a search window prominently displayed on every page of every portal. You can search by title, author, ISBN, publisher or key word, and find the book you already know you want.

Even better, at online portals, the book is increasingly more likely to be in stock and available. If the book is not in stock or even if it’s out of print, online portals typically provide links to used copies available at other vendors. This saves book fiends like me a ton of time and footwork. With portals like Amazon offering free shipping to consumers who spend a specific amount, shopping online for a specific book is actually easier than going to a bricks-and-mortar bookstore.

But what about those impulse buys? What about the books you didn’t even know you wanted? For many people – like me – those books vastly outnumbered the ones I knew I wanted. I could go to a store seeking one book and come home with fifteen. Online portals have been trying to find ways to sell those additional books, which means they have to understand and replicate impulse shopping.

What drives impulse shopping? What helps you to find a book you don’t yet know you want?

2. Position
When you visit a bricks-and-mortar bookstore, you must enter the store through the front door. This sounds self-evident but think about it. For decades, publishers have spent a lot of money ensuring that certain titles are displayed prominently at “front of store”. You know what it’s like to walk into the store and see an entire wall of the newly released (just for example) John Grisham book. This is paid space, and it works – we tend to buy what’s presented to us. If you had any interest in a John Grisham book, even if you didn’t know there was a new one out, this display might prompt you to pick up a copy and read the blurb on the back, maybe buy it. If you had no interest, the size of the display might make you think you were missing something, and propel you to check it out. The cover might catch your eye, or the title, or the display might just register as yet another mention of the book in your world.

Similarly, at grocery stores and drugstores, books are displayed near the check-out counter, along with magazines. This is because we browse them while standing in line, and often toss one (or six) into our order. There is a marketing theory that seeing a product ten times in fairly rapid succession can prompt a consumer to buy it, or at least consider buying it. So, if you saw that book in the bookstore, a billboard on your way to work, a display in the drugstore, an ad on the subway, heard an ad on the radio, etc. etc., over the course of a week or so, you might have a closer look at it, and even buy it.

Even back in the genre section of the bookstore – if you’re buying romance, the romance section is always at the back of the store because romance readers are the ones who will walk that far to get what they want and not give it up en route. Yes. They do foot traffic studies to determine these things! – you will find certain titles prominently displayed in key positions. They might be on the end cap. They might be faced-out (instead of spine-out). They might be in a dump or lined up at eye level. This often is also paid space.

There’s a huge difference with online shopping because a consumer doesn’t have to go through the home page of the portal. You can follow a link from an author website directly to the product page for that author’s book, or to the author’s page at Amazon. You could bookmark a bestseller list (maybe for Gothic Romances) and hop directly to that each time you visited the online store. You could go directly to your shopping cart or your wish list. This means that front of store displays in the virtual world are not nearly as effective as in bricks-and-mortar stores – the vast majority of customers may never see them. You might not even see the virtual end-cap in the appropriate section. And you certainly don’t have to line up and wait to check out, which means displays at check-out don’t work either.

What do you see while shopping at an online bookstore? Once upon a time, if you went to the product page for a book, you saw that book. No more. Now, there are lots of side banners and scrolling displays on every product page, and they change each time you look at the page – they change in real time, based on sales results tabulated in real time. They also change for different customers, based on the shopping history of that customer. On any given product page on Amazon, there is a horizontal display “Customers Who Bought This Item Also Bought…” with a number of titles. Below the product information is a second list “What Other Customers Do After Viewing This Item” and at the bottom of the page is “Customers Who Bought Items in Your Recent History Also Bought…” Just glancing at one of the product pages for one of my books, there is room for about fifteen other titles to be displayed on that product page. (If I were using a larger screen, there would be more.)

Notice the big difference here, what Mr. Math calls “the democratization of popular culture”. These are not endorsements from any official organ and they are not usually paid space: the online portal is telling you what other customers (like you) purchased. That’s a huge change. We are tribal creatures and we like to know what other people are reading. We’re curious about whatever has caught popular attention, and that works with books, too. Following similar logic, reviews from readers on Amazon are displayed prominently, sometimes as prominently as those from official review publications. We now tell each other what we like to read – or the algorithm does it for us – instead of paid book reviewers telling us. We’ll dig into that nice meaty topic on another Thursday. For now, let’s just take note of all these mechanisms that are trying to help us add to our shopping carts.

3. Bestseller Lists
Because we are tribal creatures, we also refer to bestseller lists. If a book is hot or popular, we will want to have a look at it, maybe even read it. The mechanism of bestseller lists works exactly the same way in every format – bestseller lists are self-propagating. What that means is that once a book lands on a list, it tends to stay there for a while. Clearly, some people only shop by referring to the bestseller lists. It seems sometimes that the majority of buyers do that.

Traditional bestseller lists include the New York Times list of bestselling books and the USA Today list of bestselling books. What distinguishes these lists is that they tend to overlook genre fiction to a certain extent. They also are not necessarily based on raw sales data. For years, there was no point-of-sale data for books: now that there is Bookscan for point-of-sale data, there are still major retailers that opt out of providing their data. (Like Walmart.) Whenever there are gaps in the data, people will try to extrapolate – how they extrapolate reveals their assumptions.

Online bookstores don’t have any issues with omission of data. They are, after all, referring only to their own sales data in the creation of their bestseller lists. There is no need to extrapolate. (There can be, however, a desire to manipulate. We’ll get to that another time.) So, you can go to the relevant section for any genre of fiction and find a bestseller list for that niche. Online portals improve the bestseller function by having more specific lists – you can, for example, look at the top 100 bestselling time travel romances on Amazon. In a physical bookstore, you’d be peering at the spines, alphabetical by author for all romance, or asking a clerk for help in finding time travels.

The result is the same, though – many people shop from these lists. Once a title is on a list, it might stay there for a long, long time. The shift to consumers driving the results is also visible in these lists – they are reflections of sales, which means they are created by what each of us buys.

4. Recommendations
This is one really juicy contributor to sales that online vendors are trying to replicate. Once upon a time, your bricks-and-mortar bookstore might have a section manager – a romance section manager – or a specialist or an employee who read avidly in the genre you prefer. Since that person might not be there when you came to shop, physical bookstores created shelf-talkers. Those are the little cards that say “Amy loved this book.” or “Amy says if you love Diana Gabaldon, you don’t want to miss this debut author!” The bookstore might also have a newsletter or produce a flyer for avid readers in a particular genre. Such a publication would likely include paid space (essentially ads from the publisher for a specific title) as well as the recommendations of the resident genre expert.

If you like Author A, you’ll love Author B. This is the nutshell of recommendations. The issue has been recreating that function automatically.

There have been a couple of ways for this to happen online. One remains direct word of mouth: you and your friend have similar tastes so you share books. That’s the oldest kind of referral and it will endure forever. Goodreads takes this to the next level: you can befriend other readers on Goodreads who have the same taste as you, no matter where they live or whether you’ve met in person. Goodreads is a fabulous network for recommendations – I can validate someone’s review with respect to my own taste by browsing their list of read books and seeing how often we agree. That’s terrific stuff for those of us on the hunt for new authors. It’s no surprise that some portals (like KOBO) have an arrangement to import Goodreads reviews and display them on the KOBO product pages. These endorsements can drive sales.

The new tweak to referrals is a personalized recommendation based directly upon your shopping history. This reveals itself in “You Might Also Like” options at check-out, or emails that come from the online shopping portal. Initially, these were kind of goofy because they were based on very little data. (I remember the eBay ones: “You’ve just bought this Singer buttonhole attachment. You might also be interested in these yellow rubber boots.” Really?) Over time, they have become much more sophisticated and more likely to be right.

In addition, instead of a bookseller realizing that Author A and Author B appeal to a similar audience – an observation that relies upon the reviewer reading a lot and thinking analytically about fiction – online vendors have the shopping history of millions of customers. They can easily tabulate the various Author B’s whose work is routinely read by fans of Author A – and they do. In this regard, online vendors are rapidly becoming better than your local section manager or genre expert in making recommendations that are right and getting you to buy another book.

5. Browsing
This is the one buying pattern that remains elusive. How would you mathematically replicate the process of cruising a shelf of books, or strolling past a display, and having a cover catch your eye? If you shop like me, the book in question might be out of your usual reading genre, it might just have an appealing cover, it might be an author you remember having recommended to you once upon a time…there are dozens of variables, none of which are readily revealed by my shopping history or viewing at an online bookstore.

I do think we’re going to see more random suggestions on product pages or customized displays for consumers in order to try to claim this share of the market. (They probably should pitch books with red covers to me.) Until then, this remains the Holy Grail to be replicated.

Maybe next time we should talk about The Algorithm.

So, what do you think? Do you buy books differently now than you did before? What elements do you like better? What ones do you miss?