My first two posts (here and here) on personalized predictive media covered both Amazon and Findory as examples. Both services give to me content not only that I know I want, but also content that I wasn’t even aware I wanted. Another illustration of this personalization/prediction in effect is offered by Tivo. Like Amazon’s purchasing recommendations and Findory’s news/blog suggestions, Tivo records and offers users television shows based on their past viewing patterns and user input.
To be honest, I haven’t found Tivo’s suggested programs to be particularly accurate or useful to what I actually want. I am not sure if they are still working on their engine, or if my television tastes are difficult to discern, but my satisfaction with this aspect of their service has been lacking. Have others had similar experiences?
That being said, I raise the Tivo example for two reasons. First, because it is a high-profile company heading towards a personalized predictive media offering. Secondly and more importantly, Tivo seamlessly integrates the personalization/prediction directly into the user experience. On the Tivo remote control are thumbs-up and thumbs-down buttons for the user to directly input her input. The engine-determined content rests side-by-side the user-requested content in the listings area, the “Now Playing List.” The key here is that the personalized predictions aren’t a separate component, but rather play an integrated role in the user experience. When a user isn’t required to think about ensuring that content finds them, and does anyway, it is then when it is most valuable.

Yesterday I posted about how I believe the consumption of media is trending towards becoming both personalized and predictive. I not only want to listen/read/view media that I know I want, but also want have media served up to me that I don’t even know that I want.
The RSS protocol provides me the first step towards personalizing the content I receive. I can subscribe to any blogger, podcaster, or news site feed that I desire. Even better, I can set up a NewsGator Smart Feed to monitor a certain subject area or keep updated on what’s being said about a specific company. With RSS, I receive the information how I want it, when I want it. And that I really like.
Yet, for the most part, I need to actively seek the content that I desire. What I really want is the content to find me.
Findory, though, is a company that is pushing forward with a vision of delivering content that is both personalized and predictive. For both news and blogs, the company’s service recommends content based on what I’ve read in the past. In their words,

“Our personalization algorithm combines statistical analysis of the article’s text and behavior of other users with what we know about articles you have previously viewed.”

So as I continue to read my Findory RSS feed, the better it becomes about predicting what content is most relevant to me. I’ve been using this service for a few months now, and it has grown to learn that I like news articles about blogs, technology, and venture capital. In short, Findory allows the right article to find me, as opposed to me looking for the article.
Interestingly, earlier this week Findory launched its personalized advertising engine. So not only is the company serving up content that’s personalized and predictive, but it’s attempting to do the same with advertisements as well. In the words of founder Greg Linden,

“Just as Findory’s personalization engine matches content to interested audiences, our personalized advertising matches advertisements to interested people. After all, at its best, advertising is a form of content. It is useful when it is relevant.”

(John Battalle offers some additional color here.)
When advertising becomes both personalized and predictive, it actually becomes content – advertorial content. At Sombasa Media, the company I previously co-founded, we were heading towards this vision before we were acquired. As our archived home page and technology page note,

“Sombasa transforms marketing materials into content that is both anticipated and welcomed by its recipients… [The company’s] proprietary technology matches merchants’ products to a specific user’s profile, determining how “appropriate” each item is for a particular subscriber. With a list of matches between merchandise and a user’s tastes, Sombasa’s technology generates and delivers a unique and personalized e-mail publication for each recipient.”

Yes, that was five years ago and our venue was e-mail instead of RSS. We were making only the first steps towards a vision of personalized predictive advertising. Findory, however, is now making much longer strides towards both personalized predictive content and advertising. And I believe that is the future.

I love Amazon.com.
Why?
I buy a lot of things from Amazon, but mostly music cd’s. There are three reasons for this:
1. Subscription shipping. With Amazon Prime “all you can eat” express shipping for $79/year, I never have to worry about shipping costs. I don’t hesitate when purchasing something, even if it is a small order, because I know shipping costs are already covered.
2. Long tail product offering. The music that I listen to and purchase isn’t always in the mainstream, so it’s difficult to find. I usually can’t find it in Best Buy or other physical retailer, nor even on iTunes. But it is nearly always on Amazon, which is great. It’s only rarely that I am looking for something that I can’t find.
3. Personalized recommendations. I have purchased so much music from Amazon that my preference profile is very rich. So when they recommend via e-mail or on their site that I purchase an artist’s album that I haven’t heard of, I’ll check it out. I often buy it – and like it. The service’s recommendations now are so spot on, that I truly trust it. That’s a powerful thing to say: that I, as a consumer, trust that when a company says I should buy something from them, I do.
I bring up the case of Amazon example as just one example of where the consumption of media is headed. It will become personalized: it’s delivered just for me and no other consumer. And it will be predictive. It’s not just about giving me what I want and only when I want it, but also telling me what I want before I know that I even want it. Media should know where to find me; I shouldn’t have to know where to find it.
So it’s more than just personalized content. I think the power of personalization resides in telling me what I should be looking for. Amazon is already headed in that direction and is doing a great job. There are other examples, too, in the RSS world that I’ll touch upon in subsequent posts.

Traditional radio is really feeling the heat these days. Increased digital downloads (legal and illegal), satellite radio usage, and IP radio alternatives (including podcasting) are consuming a larger share of people’s music listening time. Radio is slowly losing the power that it once had, as evidenced by the following ad that I found while flipping through a magazine:

“Tons of artists. Zillions of songs. And oh yeah, it’s free. Radio. Brought to you by America’s 13,000 local radio stations that play artists like these every day.”

Are ads like these going to have the same rejuvenating effect on the industry as the “Got Milk?” ads did for its own? I hardly think so. It appears to me more like a desperate attempt to grasp whatever still remains before these digital technologies facilitate change in a significant way.

With yesterday’s news that Friendster is laying off people and that the company’s CEO will be leaving in a few weeks, it’s apparent that some of the first generation social networking sites are hitting a few bumps in the road. Bill Burnham wrote about this problem in a great post a few weeks ago,

“You see, despite all the hype about social networking, it has now become readily apparent that social networking is not an application in and of itself, but rather a by-product of other activities.”

In other words, there needs to be a reason why people are getting connected. Jeff Clavier continues along that mode of thinking,

“The first generation of social networking sites (Friendster, Tribe, ZeroDegrees, Orkut, …) have all gone through ups and downs (more downs) as they were pioneering in this new space – and not really figuring out a business model for themselves, besides advertising. Social networking is now an integral part of the fabric of Internet applications, but offers limited value in its own right – with a very quick decay of one’s interest.”

I would argue that in addition to possessing a reason d’etre, successful social networking companies will more closely integrate the revenue model into the functionality of the service. It’s not just about throwing up some advertising. Take, for example, H3, which embeds the purpose of the network (locating job candidates) directly into the revenue stream (a bounty for a located candidate). I think that we’ll continue to see closer alignment of the connections’ goals with the revenue derived from them.

We all know about search engine spam. Wikipedia defines it using the coined term “spamdexing” as,

“the practice of deliberately and dishonestly modifying HTML pages to increase the chance of them being placed close to the beginning of search engine results, or to influence the category to which the page is assigned in a dishonest manner. Many designers of web pages try to get a good ranking in search engines and design their pages accordingly. Spamdexing refers exclusively to practices that are dishonest and mislead search and indexing programs to give a page a ranking it does not deserve.”

To combat the numerous techniques used to spam search sties (like keyword stuffing, invisible text, cloaking, etc.), the engines have deployed a variety of algorithms to determine ranking relevancy. (In a post earlier this month I talked about the fine line between search engine spam and content, arguing that different parties would disagree as to what is spam and what is actual content). Thus far, the major search engines have done a fairly (but not perfectly) good job at combating this problem, but it’s obviously a continuous battle.

Now enters the world of RSS and the Incremental Web. No longer is the position of a search result a function of its relevancy, but it is also a function of its timeliness. Consequently, search engine spammers have a new trick to play with.
For example, I’ve noticed as soon as I ping Technorati with a new blog post, a search for many of the keywords in that entry places my blog in the first result. As the day progresses, the blog entry moves down the list of results. I believe that spammers will begin to realize this “opportunity” to instantly have their pages placed at the top of the list results and exploit it.
Eventually the Feedsters and Technoratis of the world will determine algorithmic techniques to combat this problem, but I predict that there could be a bumpy transition period as spammers realize the power that’s here.
UPDATE: Since writing this entry, I’ve come across two great posts (here and here) from the hyku blog about spam moving to RSS and another with PubSub’s thoughts on the issue.

After the extended media blitz in the past few weeks, along with the recommendations from fellow bloggers like Seth Levine, I decided to pick up a copy of Levitt & Dubner’s “Freakonomics: A Rogue Economist Explores the Hidden Side of Everything.”
While I am almost certain that this book is going to become the over-hyped over-fashionable cocktail-conversation book-du-jour (think: Gladwell’s Tipping Point), it is well worth the read. Seth summarizes,

“The basic idea of Freakonomics is to use statistical analysis to explore relationships and answer some pretty interesting questions about our world (are swimming pools more dangerous than guns; why do drug dealers live with their mothers; how can we tell if sumo wrestlers cheat; etc).”

I was happy to read that Levitt & Dubner directly address (more articulately than I) the relationship between correlation and causation that I passionate wrote about in my post, “My Pet Peeve.”
Page 163 reads,

“A regression analysis can demonstrate correlation, but it doesn’t prove cause. After all, there are several ways in which two variables can be correlated. X can cause Y; Y can cause X; or it may be that some other factor is causing both X and Y. A regression alone can’t tell you whether it snows because it’s cold, whether it’s cold because it snows, or if the two just happen to go together.”

Of course, I would take issue with many of their arguments and conclusions. For example, they try to dispel the conventional wisdom that flying is safer that driving, “The per-hour death rate of driving versus flying is about equal. The two contraptions are equally likely to lead to death.” I ask: isn’t the more relevant metric the per-mile death rate? Afterall, there’s a reason that people fly – to get there faster.
But more importantly, one of their basic arguments is that so-called “experts” in a field (like real estate agents) are often incented to not necessarily present the full truth. And these “expert” authors here run into the same quandary. With strong financial motivations to sell more copies of the book, Levitt & Dubner are compelled to make wild claims, which they do: the decline in the national crime rate in the nineties was a consequence of Roe vs. Wade; parents’ actions have little effect on the outcome of a child. There potentially may be some evidence to support these extremely controversial claims, but we must remember that the motivation of these authors comes down to the same principals which they themselves argue – economics.

While we transition from the Reference Web to the Incremental Web, more and more information will be available online – especially information about you.
As the web moves from merely a reference medium to a true conversational medium, and the tools to post our thoughts and digital content becomes more accessible, the amount of information about individuals will increase. And search engines dedicated to find that information will also flourish.
I think everyone (or at least nearly everyone) has “Googled” themselves to find out what is posted on the web about them. Yes, but have you “Technorati’ed” yourself lately? Other than professional bios, the reference information available when I’ve Googled my own name is pretty basic (for example, a quote in the New York Times and the results from an Angel Island 12K race a few years back). And my Technorati results are mostly just my own blog entries & trackbacks to them.
But what happens when your online reputation goes awry? In the course of my job, I was doing due diligence on one nameless entrepreneur and learned his fiance’s “pet” name for him in her blog. That, I am sure, he didn’t wish for someone in a professional context to see. Embarrassing, yes, but innocuous as well. Yet you can see where this could go if only taken one step further. An angry ex-spouse rants on his/her blog about the other, or a not-so-flattering picture from last year’s company holiday party tagged with your name on flickr. That could do some actual damage.
Your online reputation does matter, and I’d argue it will increasingly do so.
For example, when Fred Wilson of Union Square Ventures hired Charlie O’Donnell, he turned to online research, “We found his blog to be the single best diligence item in our process of hiring him.” Now that’s a powerful demonstration of a positive online reputation.
Increasingly, we are going find out information – both professional and personal – about people online. (I know from my own web server logs that ten people found have found this blog so far this month by typing my name into one of the search engines.) And as more information about us is online, people will be increasingly likely to search for us.
Do you know your own online reputation? It’s out there whether you know it or not.

In a recent Business 2.0 interview with Om Malik, Cisco Systems’ Chief Technology Officer Charlie Giancarlo is quoted as saying,

“I think 100 megabits per second into the home is a given [in the near future] if we want more HDTV channels on our IP connections. Inside the home, 1 gigabit per second is going to be necessary as we move those big video and audio files around. In a decade or so, I expect that gigabit wireless would be cheap enough as well.”

The only question that I would ask is: with each member of a four person household all simultaneously using utilizing HD IPTV, VoIP, file sharing to multiple devices, automated storage backup, and many other imminent bandwidth intensive apps – perhaps we need 1 Gb/s into the home sooner than in a decade?
There’s no question, though. It’s coming.

The CEO of Indeed, Paul Forster, addressed a number of topics on yesterday’s Syndicate panel entitled “New Pathways to Profits: Exploring Business Models.” I caught up with him after the session to chat about the very insightful distinction he made about how corporations can monetize content, even though RSS allows it to be decoupled from its origins. After all, how do you monetize content when it’s everywhere and no longer under complete control?
Paul drew a distinction between what he calls explicit monetization vs. implicit monetization. Explicit monetization is direct and overt. By definition, user-generated revenue is tied directly to the feed itself. The best example is the placement of ads in publisher content RSS feeds. Classic model of content surrounded by advertising.
Implicit monetization is more subtle. The way I understand his characterization, corporations can offer RSS feeds which are part of the value of a user’s revenue stream, but the revenue generation isn’t directly tied to the feed. For example, UPS or FedEx offering a tracking feed tracing the path of a customer’s package. It is not explicitly a revenue-generator, but rather delivering service value which the customer certainly pays for in turn.
I would also argue that there is a third category of monetization somewhere in between explicit and implicit, which I’ll call advertorial monetization. There is a distinctive set of content that companies can offer which is both editorial content and advertising – i.e. advertorial. In these feeds, the content is the advertising, yet it still delivers value to the end-user. For example, e-retailers could offer consumers a feed with weekly coupons. Or, eBay could offer a feed with the persistent results of my search for “t-shirts,” so I always know what’s for sale on its marketplace. In both of these cases, these feeds are advertorial because they are clearly merchandising a company’s products, yet they still offer value to the end-user.
I think that both start-ups and large corporations alike can benefit from exploring which of these three monetization strategies are best suited for the type of syndicated content they offer their customers – explicit, advertorial, and implicit monetization. It’s not a one-size-fits all.