Amid increasing reports that Amazon continues to dominate the market for e-reader and ebook sales, Sony and other competitors are altering their strategies. Last week, I wrote a post on why Sony had little choice but to give up its proprietary BBEB file format and instead adopt ePub and Adobe DRM. Sony’s new strategy is an attempt to modularize the market and to thereby allow several companies to succeed.

In this respect, Sony’s embrace of ePub was necessary. However, the decision is, in and of itself, insufficient if it wishes to change the market’s overall dynamics. In a recent article on Slate, Farhad Manjoo lays out his suggestions for how Sony (and other Amazon competitors) should proceed if they wish to be successful. The underlying logic of his suggestions comes from the music industry’s transition to digital and Apple’s ability to dominate the new distribution channel. He states:

Anyone looking to beat the Kindle, then, should look to the iPod: Study everything that Apple's rivals did, and do the opposite.

While ebook companies would be well advised to study and understand the history of this market, they should be careful when extrapolating the experiences of Apple’s competitors in order to formulate their own strategies. Digital music is very different than ebooks. Having said this, Sony and others should take the following lessons from the music industry:

It’s all about the experience

As I have said before, consumers do not simply buy products. They do not merely buy content. Consumers buy experiences. With ebooks (as it was with digital music), consumers have paid little attention to the file format or the DRM protection on their content. Now, this is not to say that no one cares about these things. Many consumers do, and these individuals tend to be very vocal about their concerns. But, these individuals are a minority. In fact, according to a recent report done Imran Khan, only 15% of consumers cite Amazon’s proprietary file format when deciding not to buy a Kindle (and only as a 2nd or 3rd reason).

Switching Costs Matter

Manjoo cites the failure (or at least the underwhelming market penetration) of Apple’s competitors as an example of why the overall ecosystem matters. While he is correct, something much more significant is occurring with the Zune and the other device manufacturers that he referenced. By the time the Zune entered the market (in 2006), Apple had solidified its position and had established a large user base. To compete successfully, Zune had to not only provide a better user experience, but it needed to convince users that the Zune experience was so much better than Apple’s that it warranted the incurrence of considerable consumer switching costs. These costs included not only the amount of money invested in music (compatible at that time with only iTunes) but also the time spent learning and becoming comfortable with the iTunes and iPod systems. Consumers like consistency, and comfort with an existing platform represents a considerable obstacle to would-be competitors.

Sony is in a different position today than Microsoft was in 2006. The ebook market is still relatively small. Only 2% of consumers currently own a device and the market is expected to grow significantly. This provides opportunities to Sony that won’t be available if Amazon continues to dominate into the future. However, switching costs are still relevant in the ebook market.

According to Carl Shapiro and Hal Varian, consumers who experience lock-in go through the following cycle:

Lock-in begins when a consumer chooses a brand, begins to sample it, and then invests in it. While we can see that current Kindle consumers have done exactly this, such a view is far too narrow in its scope. The degree of Kindle’s success is due, in large part, to the fact that consumers had previously selected the Amazon brand as their platform for purchasing products. For many, therefore, the purchase of the Kindle is simply a further investment in the platform (and an additional source of overall lock-in). Sony will need to understand this lock-in and be able to address it when it makes marketing and distribution decisions.

Align incentives

Manjoo’s idea of trading ebooks among friends is similar to Microsoft’s idea for social music on the Zune. Individuals could send a friend a song who could then listen to it up to 3 times over the course of 3 days. The concept, despite its uniqueness, did not lead to the level of success Microsoft expected because the feature’s value materialized only if a consumer’s friends also bought into the Zune platform. For these individuals to buy into the system, they would need to incur those considerable switching costs for what was a pretty insignificant benefit. Listening to a song 3 times, after all, could be achieved far more conveniently. Zune’s social media concept failed because the user’s incentives were not aligned.

Manjoo’s idea -- while intriguing -- falls into a similar incentive trap. This time, however, it is the incentives of the content creators that creates an issue. Allowing Zune consumers to listen to a song 3 times for free was allowed because a user’s demand for repeat consumption is high in the music industry. So, giving away the content (while placing certain restrictions on convenience) is effective at increasing purchases. The same cannot be said for books. Once a consumer has read a book, that individual’s demand for the content decreases dramatically. If consumers are allowed to share the full content of an ebook with friends, sharing will in most situations not lead to increased purchases. More importantly, such DRM policies could potentially open the door for third party ebook rental markets that could actually reduce revenue to publishers. The incentives for publishers are simply not aligned with this proposal.

Manjoo’s idea has the correct end goal -- distinguish the ePub platform by increasing consumer options. However, ebook aggregators and device manufacturers must pursue strategies that have incentives aligned for all interested parties.

Consumers have more choice than ever in the book industry. According to Bowker, there were more than 275,000 books published in 2008. That’s approximately 1 book every 2 minutes. With all of these choices, it can be difficult for authors to find new readers. And, while previews of content undoubtedly help, many have discovered that free content works even better. When I checked the Kindle store this afternoon, I counted 4 books in the top 10 that were free. The popularity of these ebooks has received a lot of attention both in the media and among bloggers. But, does free really make sense in the ebook world?

Business models based on free are nothing new and can be seen throughout the economy. These past and current practices offer good lessons to publishing companies and self-published authors trying to navigate the nascent ebook market. These models, despite their many obvious differences to one another and to ebooks, have an important commonality: to be successful, free must lead to paid.

This link can be accomplished in two ways. First, free can overcome the consumption hesitancies caused by purchasing information. Books (and other information products) are considered experience goods -- consumers cannot truly assess their willingness to pay for a book until after they have read it. Companies have traditionally addressed these hesitancies by giving away a portion of the content so that consumers can better estimate its value. Amazon, for example, lets ebook readers have the first chapter of any title for free.

The second way in which free can be successful is if it increases consumption of complementary, paid goods. The newspaper industry provides an excellent example of this strategy. For all of the problems the industry is currently experiencing, many companies are able to generate positive contributions from their online divisions by giving away content. This model works because free content attracts readers. And, the more readers a company has, the more click throughs the company will receive on its advertisements. On newspaper websites, content and advertisements (if designed well) work as complements. Users consume both together.

Let’s look at each of these possibilities in turn:

Consumption hesitancies.

Allowing for partial samples of a book helps to reduce the fears and uncertainties associated with purchasing information. However, more recent arguments surrounding free focus less on partial product distribution and increasingly on total product give away.

The strategy of free digital distribution worked fairly well in the early days of the Internet. Authors could post content on the web and see an associated increase in print sales. Such an increase was possible because consumers (after evaluating the content online) still wanted to purchase the more convenient, print version. The value of such a strategy today, however, is uncertain. Many users seem comfortable reading on an LCD screen and those with e-ink devices can easily transfer content to eliminate any inconvenience. These changing consumption preferences impact the viability of this strategy because books are a unique type of information good (distinct from music or movies, for example). With books, there is little demand for repeat consumption. Once a user has read the book, his/her demand drops dramatically. The connection between free and paid, once driven by format convenience, is now lost.

Because of this, free’s ability to increase profit by reducing consumption hesitancy depends on its ability to generate positive feedback in overall consumer purchases. There is little doubt that a book’s strong sales today impact its sales tomorrow. A successful book gets more visibility, and visibility translates into additional purchases. Free could theoretically help to accomplish this: publishers could release a book for free and begin charging for it once it became sufficiently popular. This seems to be the strategy that Chris Anderson employed with his recent book. The title was available free of charge for a short period of time through the Kindle store. In that time, the book earned a place on the Kindle’s Top Seller list. Soon thereafter, Anderson’s price jumped to $9.99.

While this strategy makes some sense, it suffers from the fallacy of composition -- the idea that what is true for the part must also be true for the whole. In this instance, a strategy of temporarily free titles to create positive feedback is effective only if others do not also employ it. Each additional free ebook on the market would diminish the visibility of other free titles, reducing any individual book’s benefit.

Complementary Paid Goods.

Publishing companies and authors are now beginning to offer the first book in multi-book series for free in order to maximize overall sales. The free first book reduces consumption hesitancy and increases demand for the complementary, related titles. Based on statements made by publishing companies, this strategy of free can be very successful.

The problem with this model, however, is that it works only in very limited circumstances. Its success requires multiple, related books (ideally in a series). However, to get multiple books published (at least by an established publishing company), an author has to have already experienced considerable success. In fact, of those top 10 books that were free on the Kindle store, all came from proven, well established authors of fiction.

Free can work for ebooks -- but only in very limited scenarios and for a small subset of authors.

A few weeks ago, I wrote a post analyzing whether ebooks cause print cannibalization. The motivation for the post was Sourcebooks’ decision to delay the ebook release for “Bran Hambric: The Farfield Curse” in an effort to prevent hardcover cannibalization. A similar issue has come up again (albeit with a different result) with Dan Brown’s new book, “Lost Symbol.”

The issue of cannibalization is important for publishers and centers around the practice of versioning. For-profit companies design and implement strategies in order to maximize profits. In theory, profit would be maximized when a company is able to charge consumers their individual willingness to pay for an item. However, such personal knowledge is not easily shared by consumers, making this theoretical pricing practice impossible. Versioning is an attempt to understand consumer’s approximate willingness to pay and charge accordingly. Companies will release one version of a product designed to appeal to high paying consumers and more basic versions targeted at others.

Companies can implement versioning strategies in several different ways. Publishers, however, have long used time versioning. Publishing companies attempt to identify those consumers with a high willingness to pay by initially releasing only the more expensive (and more profitable) hardcover version of a book. Those who place the highest value on the content will want the book immediately and will not wait for the lower priced paperback. The result is higher total profit.

The fear among some publishers is that the simultaneous release of hardcover and ebook versions will limit the effectiveness of this strategy and thereby reduce profit. However, the underlying assumption of this strategy -- and the reason why this type of versioning works -- is that consumers choose content first and format second. Ebook consumers seem to be challenging this (at least those who have purchased a dedicated device).

Choice of format before content clouds a publishing company’s ability to use time as a metric of willingness to pay. Therefore,while ebooks will likely have a limited impact on a company’s print versioning strategy, publishers will need to find new methods of segmenting ebook consumers. Such segmentation can still occur through versioning strategies, though companies must be willing to experiment with different applications.

Random House, for example, has launched Book and Beyond in the UK. Consumers most interested in an author’s content can pay extra for enhanced versions containing audio and video components, interviews, and images. Those with a lower willingness to pay will likely forgo these extra features and purchase the basic version.

Other types of versioning strategies exist and could help publishers segment the ebook market. However, the success of these strategies may be determined by the extent to which the ebook market becomes modularized. Complete, vertical control over a market has its advantages, but it often acts as a disincentive for experimentation.

Publishing companies should recognize that current versioning strategies are effective (and therefore profitable) because they allow consumers to reveal their willingness to pay. While ebooks challenge current versioning practices, the same result can be obtained if companies are able to experiment with new methods.