The Building Blocks of Personalized Programming

Hundreds of television channels, thousands of Websites, and millions of videos are all a few clicks away. We have entered an era where there are very few barriers to making content available to consumers, leading to an unprecedented amount of entertainment options from a rapidly growing number of sources. This has great promise because of the high likelihood that something finely tailored to every consumer’s taste is available out there somewhere.

Sounds great, right? In theory, yes -- in practice, not yet.

Here’s the problem: As consumers, we have been trained to lean back and accept the entertainment programming that has been put in front of us. This is changing, to be sure, with consumers shifting their viewing time and place. We are leaning forward a bit, taking control of our entertainment, but mostly to interact with the most popular, pre-programmed content. To go beyond this, consumers have to either lean way forward and search for content, which implies that they know what they are looking for, or browse, which takes a long time and may not pay off with content they find interesting.

Programming has a valuable place in the entertainment universe. On TV, a network orders its shows to naturally lead the viewer from one program to the next. This provides convenience and a sense of flow that can make the experience more enjoyable with less work. The challenge in front of us is how to apply the notion of programming to a much broader set of content, while presenting a very personalized experience.

Clearly, the traditional model of people making editorial decisions that result in a broadcast schedule begins to break down as the model moves increasingly towards a 1:1 engagement with consumers and includes a much larger body of content. Technology will have to play a bigger part in determining what is presented to a consumer if the promise of personalized programming is to be realized. What’s the payoff for figuring this out? The payoff will be a better consumer experience, driven by the presentation of more relevant content.

While there is no silver bullet that addresses this challenge, the right building blocks are in place to make personalized programming a reality. One theory gaining momentum is that if programming is informed by multiple sources of information, the odds of increasing relevance go up significantly.

There are three, maybe four, interesting buckets of information that lend themselves to maximizing relevance. The first bucket is information that content owners provide about their video, such as title, actors, genre, and a descriptive brief. This is the primary source for programming guides today.

The second is information that can be derived from the content itself. This relies on technology that converts speech to text and, in the near future, will recognize people, places, and things in the video. The tricky part is contextualizing what is derived from the video into meaningful data that can better inform the programming. This technology is improving, resulting in a video timeline that highlights key topics and points of interest.

The third bucket of information is derived from consumers’ rating, tag, and recommendation data, which can be used to learn more about the content and the consumer’s relationship with it.

The last bucket is behavioral profile information, which is collected by tracking a user’s Web searches, sites visited, and products purchased. Because the required technology is beyond the capability of most sites, and the privacy issues surrounding such data gathering have yet to be vetted, it is unclear whether this last bucket is worthy of pursuing at this time.

If you layer better-informed programming with technology designed to improve and measure the effectiveness of the content guide, the results are very smart, proactive recommendations that give consumers the convenience of low-touch programming with the benefit of highly personalized entertainment. Use this approach effectively and everyone wins. Consumers get a better experience, video sites increase their audience, and, ultimately, we can apply what we’re learning to television. Sounds great, right?

For content companies interested in this vision, there are a few things to prepare for as we wait for the technology to catch up. Invest in an information architecture that will access data about your content. A thoughtful "metadata schema" applied across your organization will enable you to capitalize on the latest programming tools. Make sure this information is archived in an easily accessible system. The ability to retrieve this data will ensure that you can present your entire catalogue of videos.

Finally, review your company’s privacy policy with an eye on using information like comments and ratings. Most policies were written before the explosion in social networking and didn’t anticipate applications like a "Highest Rated Video" widget for Facebook or MySpace . With these simple tasks completed, you should be well positioned to expand your audience through personalized programming.

— Ian Blaine, Founder and CEO, thePlatform