I realized that I have been spending a lot of time thinking about why some content rental markets are better developed than others – both in terms of content type and geography. As I thought deeper, I started collecting enough data to put my thinking into a framework-like approach, where I analyze the tendency of a consumer to either OWN or RENT media content. Below, you will find the first version of this analysis, which will probably be subject to substantial improvements in the near future.

The major types of content discussed here are Books, Textbooks, Music Albums, Movies, Video Games and Software. Please note that, for some types, the sub-types might have different rental potential. I will discuss those in the next part of the analysis. For example, under the Video Games heading, we need to analyze the following pairs differently: PC vs. console games. Online-enabled vs. Offline-only games. However, I will just focus on 6 aggregated content types for the sake of simplicity in this post.

The first step is to explain the factors, or determinants, driving an OWN or RENT decision for the consumer:

1. MSRP: The cost of a new release for each content type. As price goes up, rental becomes more attractive for the consumer.
2. Time to consume: The time it takes to consume the content fully. A popular book takes 2-3 weeks to read fully on average. A movie takes 2-3 hours in contrast. Students use a textbook for a full semester, which is 3-4 months. In that regard, a music CD is considered to be used for 3-5 years before it is archived or lost.
3. Re-consume value: To what extent consumers re-read a book, re-watch a movie, re-study a textbook, re-play a music cd or a game. We rarely read a book or watch a movie twice. But we play the same music album every week and use Microsoft Excel every day. If consumers re-use the content less, they would be inclined for rentals rather than ownership.
4. Copy protection: How hard it is for the consumer to make a copy of the content. It is extremely easy to rip a music CD, while it is almost impossible to make a practical copy of a textbook. Copy protection is an important factor for the consumer as they will be inclined to rent more if they can rent, copy and return the content instead of buying it.
5. Long-term relevancy: This is tricky. At first sight, you might think that this is basically the same as #3, "Re-consume value". But it is not. Think of this as how relevant a particular content will be in the future. For popular books, we rarely see big changes in the new editions. In most cases, there is a new foreword or an appendix in the newer editions. For textbooks, the change is substantial. Even though only a few pages and practice problems are added, these additions change all the layout and page numbering of the books. Therefore, students usually find it hard to follow a university course with an edition older than the one suggested by the professor. For movies and video games, situation is different. Both types of content are heavily new-release driven. Almost all titles in these two buckets generate 80% of their lifetime sales in the first-month release window. Especially for video games, online gaming is only relevant when there are other players logged in. Therefore, once a sequel is released, it is very hard to keep the older version of the game relevant. For software, a newer version of a suite takes longer to migrate, but it eventually happens. It is hard to find people using Excel 98 today.

After defining the factors, we will need to rate each content type across them. Here is the chart:

Here, the concluding row is the Rental Price Ceiling. This summarizes how much a customer would effectively pay per day to consume a content fully. For example, if you have paid $15 to buy a popular book and finished it in 18 days (2.5 weeks), basically you paid $0.86 per day. Similarly, you buy a movie DVD for $15 and in only one day, you finish it. Therefore, your effective cost per day is $15.

This metric is quite important, because, among other things like re-consume value, this calculation is the ceiling for rental opportunity. As we see here, movies offer the highest rental price ceiling with $15. At the second place, we see video games at $3.43 per day. The video games are four times the price of movies, but it takes a lot more days to fully consume a game. On the third rank, we see textbooks with about $1 per day rental ceiling.

The first conclusion is, in fact, powerful. We see the highest rental activity in movies, driven by companies like Netflix and Blockbuster. Video games are on the second place, which is perfectly correlated with GameFly's popularity in the market. On the tihrd place, the textbooks, we see a fast-growing market driven by heavily VC-backed start-ups like Chegg and CampusBookRentals. Interestingly, beyond this point, we don't see important companies serving rental needs for popular books, music and software. Music is easy. With $0.01 effective daily price, there is no economical way to rent a music cd. For software, $0.16 is almost completely inhibitive for a rental market.

The second conclusion, which is more subtle, could even prove to be more powerful. It is about what it takes to create a rental market or enhance its potential. Think about the following conclusions derived from the table:

• Popular books are almost at the same rental price ceiling as textbooks. So, why not to start a popular book rental service? First, target consumers for textbooks, the students, are more cash-strapped than the overall target group for books. Therefore, a rental market for textbooks is more attractive than for the popular books. Second important factor is the ratio of rental price to shipping cost. Think about the differences in the rental transactions in two content types. For textbooks, a students rents one and keeps it for 3-4 months. Let's say he pays $20 for that. The book gets shipped only once during this period. For the popular book, the renter would pay around $3 to rent and the book gets shipped once every 3 weeks. In that regard, the shipping costs become inhibitive for the rental company. This brings us to another opportunity. Why not use a kiosk rental model for popular books to shave shipping costs off completely?
• For video games, there are important conclusions. The fact that a video game is online-enabled increases the "Time to consume" substantially. Therefore, a rental company might focus on serving single-player-only games.
• For any content type, fixing the copy protection issues can change the rental potential of a market dramatically.
• For other geographies where prices are quite different (e.g. in Turkey where average textbook prices are around $30), a quite different rental market potential is expected. Similarly, with the average price of console games at $120, there is an attractive rental market to tap into. By changing other determinants of the market, such as the introduction of Blu-Ray with the ultimate copy protection, a robust video game rental market can be created in many countries like Turkey.

• Comments

There are strange things happening in Facebook’s “Suggestions” box. While it has worked quite nicely in the past few months about finding some “real” friends, the suggestions started expanding into some dangerous territory.

There are a number of people that keep coming up to my Suggestions list even though I cross them out one by one repetitively. The commonalities among these people are:
- They are all female (not a single male without a common friend is suggested to me, while a lot of females are)
- They all have NO common friends with me
- They all have ZERO friends
- They all belong to the same region (this is weird, as Facebook recently eliminated regional networks)
- They all have unusually minimalist clothing. Of course, not at the level of to-the-face twitter spammers but just on the grey area. Yet, you can easily assume that they are real-life persons.

Now, compared to twitter, this suggestion-spam looks more dangerous because these people do not add me as their friends. Rather, Facebook suggests me to add them even though there is no meaningful reason for them to be suggested to me. With a lot of engineers joining the company recently, could this be an evil side project of an ex-Googler that stressed-out after spending years under the mantra “Don’t be evil”?

I need answers to the following questions to conclude more clearly:
1) Do you receive similar irrelevant suggestions?
2) Do females receive similar male suggestions with questionable profile photos?

• Comments

For 2008, my picks were these.

While I change the format slightly, most of my faves remain on the list:

• Comments

FeedFlix, a Netflix API service, says that Netflix is losing 0.3% of DVDs it is shipping.

GameFly, which gives the same Netflix-style service for video games, is unhappy about ~1.0% loss rates for the DVDs it is shipping. Probably for that reason and some others, they filed a complaint with the Postal Regulatory Commission. USPS is serving both companies, and some people in the industry say this could be a potential evidence of USPS giving preferential treatment to Netflix. While it might as well be the case, I think the more important explanation lies somewhere else.

Let's say you want to own a movie and a game. The movie you want to own is Transformers: Revenge of the Fallen, which is currently topping the Amazon.com sales charts. You can buy it for $15. Or, you can start the cheapest Netflix subscription plan for $9 a month and receive the movie's DVD. Then, you can claim that the DVD has never arrived. You can then either cancel your subscription plan or become a "legit" user for the rest of your life. Yet, in this initial transaction, you acquire the DVD for a mere $9, instead of $15.

Let's now look at the GameFly side. You want to own Uncharted 2 for Playstation, which sells for $55 at Amazon. Instead, you start a GameFly membership, cheapest one being $9 a month. The same story of claiming the DVD is lost and you net out $46 profit.

Going back to the loss rates, let's consider the total loss rate TL = PL + UL, where PL is Postage Loss and UL is User Loss, all being percentages. Postage Loss is the percentage of losses caused by postage service, and UL is the other part of losses caused by user behavior. Based on the two articles above, TL(Netflix) = 0.3% and TL(Gamefly) = 1.0%. People are talking about the alleged differences in PL(Netflix) vs. PL(Gamefly), but what about the UL side?

Let's consider the two extreme examples: If a DVD you receive from your subscription was worth $1 million, would you keep it (assuming you were still paying $9 a month subscription fee)? And in the other extreme, if a DVD you receive was worth only $0.5, would you keep that one? I guess we can safely assume that when the "inherent value" of a DVD approaches infinity, the UL rate should approach 100% too. Then, it should be perfectly rational to assume UL(Gamefly) should be higher than UL(Netflix) as average game prices are significantly higher than those of movies.

I think the perfect test for this theory could come from Netflix' internal data on loss percentages for DVDs and Blu-Rays. As Blu-Ray discs are more expensive, Netflix should have a higher loss rate for Blu-Rays, compared to regular DVDs.

If the assumption holds, what can GameFly do? In my opinion, they shouldn't search for the answer in the courtroom. Instead, they should consult behavioral economists. If I remember correctly, showing the list of "Ten Amendments" just before an exam reduces the cheating rate significantly, even though the amendment has nothing to do with the content or context of the exam. While it could be weird to see the Ten Amendment page coming out of each game envelope, I am sure economists can figure out an effective "anchoring" method eventually.

• Comments

Google Reader started offering a new sorting method for blog posts: Sort by magic.

While many tech evangelists have already declared RSS-reading dead and migrated to Twitter-style micro-sharing, I still think RSS is by-far the most effective learning tool/platform on earth (maybe on par with the courses I am taking in Stanford GSB).

There is a personal dimension for me in seeing Google Reader implementing post recommendations. About a year ago, I was working enthusiastically with two classmates to crack the personalized news problem. After putting 200+ hours into it and building a prototype, we realized that there were two most serious questions we needed to answer about our strategy before moving on:

1. How do we solve cold-start issue?
2. What happens if Google starts personal recommendations?

The first question was about people starting to use a new service (probably switching from Google Reader) and having high expectations from day 1. The cold-start issue was about the fact that we wouldn’t know anything about the user’s reading preferences until we collected his reading patterns. Therefore, during that "cold" period, the user wouldn’t get any better experience than what Google Reader offered. Plus, as he was "used to" a certain RSS-reading interface, he would inevitably have an inferior experience as all humanbeings have strong anchors to what they are used to. In fact, by spending so much time on cold-start issue, we realized that many start-ups were using social recommendations (I get recommendations from the posts Mike has read, because the engine thinks Mike has similar taste to me), because when you base the engine on social similarity, you can shrink the cold-start period close to null.

Second question was a preemptive critical thinking about what our competitive advantages would be, if any, in case Google starts serving personalized recommendations. First, they had the sizeable user base to "learn" how to make better recommendations. Second, they would already have vast amounts of reading patterns, by-passing the issue of cold-start. Third, the recommendation engine would have lots of synergies with their search-ranking algorithms. Among many additional dimensions, they clearly had the right to win.

Based on these concerns, we chose to stop pursuing this project. Since then, I have been checking out Google Reader blog with the hope to see Google start its recommendation service. Seeing that would have "verified" our critical thinking about the start-up. Today, I got the news, thru Google Reader of course. Here is the first pack of recommendations I got from 1000+ unread posts sitting in my account:

Based on my 3+ years in Google Reader, spanning 10,000s of read and starred items, I would expect a better "magic" from Google at the top ten bucket, but I am sure they will get there. With this feature, RSS became an even more important source of "learning", cutting down noise. With micro-sharing platforms plagued with 99.99% noise, self-promotions and non-value added posts, I actually feel good that millions of people are migrating from RSS to mediums like Twitter. That just puts me in front of the competition as they are spending hours and hours digging thru hundreds of 140-char posts to find 2-3 valuable items.

• Comments