Me TV: Programming Content for an Audience of One
For those whose childhood overlapped with the 80s, an indelible image is the man on the moon. I’m not referring to the moon landing -- I’m referring to the images of the man on the moon introduced by the words "Ladies and gentlemen, rock and roll" and followed by Video Killed the Radio Star. For those of us who are fans of music and video, this music video was the first of many that compelled us to say, “I want my MTV.”
Over the past several years, my colleagues and I have discussed the transformation, evolution, and disruption of the television industry at length -- in order to plan for what’s to come in the future. And, in order to look forward, we must first look back.
For decades, viewers consumed broadcast content in a strict, curated format. Traditionally, viewers had the choice of a fixed number of broadcast and -- if they had access and chose to subscribe -- premium cable channels. Every channel consisted of regularly scheduled pre-recorded and/or live programs and (for most channels) paid and/or promotional advertising. Programming was scheduled by dayparting, with episodes and their associated advertisements planned to maximize viewership and revenue. Viewers turned to the daily newspaper or periodicals such as TV Guide to understand their choices.
Over the years, electronic program guides (EPGs) replaced print guide, viewers had access to more providers (cable, DBS, phone/fiber), providers offered more channels, and appointment viewing evolved to include PPV (sports) and VOD (films). Audience measurement continues to rely on statistical sampling (for instance, one of the more prominent ratings services is purported to use a sampling of 25,000 households to measure over 100 million households).
DVR-enabled, timeshifted viewing was the first massive change to viewing behavior. And, the next disruption has already started. No longer do viewers demand, “I want my MTV” -- instead, viewers demand, “I want me TV.”
Me TV = Leanback Programming
Me TV is the ability to dynamically -- uniquely -- program content for every user, optimized for a leanback experience.
Linear programming is dead. Long live leanback programming.
The foundation of leanback programming is composed of three core concepts:
- The viewing experience is content-agnostic, driven by content-decisioning logic.
- The viewing experience is addressable, personalized, and measurable.
- The viewing experience should engage the audience to dynamically shape the linear experience.
The programming notion of content channels remains valid; however, content type boundaries (i.e., pre-recorded vs. live, programs vs. advertisement) are removed. Leanback Programming treats all content types equally and enables the publisher to determine how that content is consumed during the linear video experience.
Today, entertainment programming consists of pre-recorded content aligned on the half-hour with pre-recorded commercial breaks. Live programming is similar: it consists of live content with a predetermined starting time -- on the half hour -- with pre-recorded commercial breaks.
Leanback Programming orients itself by thinking that the consumption of video content is live (to the viewer) and is agnostic as to the content source. Content could be any combination of the following:
- Simulcast, pre-recorded content from broadcast
- Simulcast, live content from broadcast
- Live streaming from a digital source
- Pre-recorded content (program, advertisement, bumpers/interstitials, delayed/time-shifted telecast)
The Audience of One
The Me TV audience is addressable; the programming delivers a video experience that is unique to the user.
Leanback Programming should
- Collect baseline information about the user’s viewing context to influence programming, including geographic location and device (form factor, screen size, connection type, network, time of day).
- If allowed by the user, collect preferences to influence programming through analysis of shared information (e.g., accessing Facebook social identity information using the Graph API, understanding members’ viewing preferences through the Netflix API, analyzing the local iTunes library for music preferences). Preferences can also be collected directly, e.g., genre, preferred duration of content, restrictions based on ratings (TV Ratings, MPAA), and language or captioning requirements. This information is not a one-time aggregation but should be updated over time.
- Measure consumption to understand trends to gauge content engagement, popularity, and relevance. In addition, quality-of-service should be monitored to optimize the video delivery process and to identify systemic issues that could affect the video experience, e.g., buffering, latency in specific regions, poor playback of specific content that could signal issues with the encoding, or irregular usage on specific devices that could highlight issues with particular hardware.
By combining the notion that programming can be any combination of content within the context that every user can have their own unique video experience, we create a model of truly dynamic programming.
Consider this example: throughout the day, I often tune in to financial news, from live coverage and analysis of the market to pre-recorded content about specific topics. While the content is important, not every program is relevant to me.
If a publisher were to redefine this for a Me TV audience, the starting point would be to identify specific simulcast -- pre-recorded broadcast, live broadcast, or live from a digital source -- content that should be viewed at a specific time.
For the uncommitted blocks, the publisher would identify relevant content based on my preferences. As the publisher has previously queried me for my stock portfolio and interests, the content filters to a target set. After identifying that I’m viewing from Los Angeles (or Boston or New York or London), content is sorted based on relevance to my region and the popularity/engagement of the content for viewers in my region.
The burden of traditional broadcast programming is that it’s always aligned with the top and bottom of the hour. In this dynamic model, content of irregular lengths can be supported. The publisher, based on its advertising and promotion needs, can have more control over how advertising is handled. Not only can the publisher insert ads between content or at specific timecode-based cue points within a program, the publisher can adjust ad frequency, ad duration, pod position and pod duration throughout the video experience. My viewer data can even be used to target more relevant advertising.
Thus far, Me TV has been focused on a leanback experience. However, this doesn’t mean the publisher should not engage the viewer. On the contrary, the personalized viewing experience creates opportunity to selectively interact with the viewer.
While I watch my finance channel, the publisher may still want to promote new programs, whether pre-recorded or simulcast. The publisher could proactively prompt me to switch -- for a period of time -- to the new program. The publisher could use a sponsorship model and offer an uninterrupted program -- or uninterrupted viewing for a fixed amount of time -- after viewing a sponsor advertisement.
In addition, if the publisher detected that I switched devices, changing from a desktop or tablet to a mobile form factor, it may ask if my programming should be adjusted for mobile consumption by targeting programs of shorter durations and limiting playback to lower bitrate renditions to optimize for 3G connections.
Publishers can apply Me TV to any number of scenarios:
- Entertainment news programming can combine daily simulcast programming along with the viewer’s favorite actors, directors, television shows, celebrity personalities, and geography.
- Weather programming can combine live updates throughout the day with reality and fiction programs with the viewer’s geography, favorite locales (e.g., an ex-Los Angeles surfer longing for the surf report from Chicago), and upcoming trips (e.g., leisure, business, holidays)
- Leanback Programming doesn’t assume 24/7 linear content. Even a sports event can be personalized. Geography can be a simple factor, but publishers can allow the viewer to choose their team, influencing the selection of the audio commentary and camera feeds -- from fans to locker rooms to sidelines -- that they have access to during the event.
- Parents could even pre-filter a set of content targeted for their children.
- For fans of game shows, programming could aggregate content by show, aggregated show clips (e.g., Final Jeopardy), or even intersperse video content with sponsored interactive games.
- For those of us that miss music videos, the visual storytelling can deliver a more captivating experience than just audio -- think “Thriller,” “Runaway,” or “Same Love.”
- Broadcast programmers haven’t shied away from themed content programming, from Shark Week to Halloween Horror nights to TCM’s 30 days of Bond. The Stanford Theatre on University Avenue still runs summer classic films programming from Hitchcock (Notorious) to Cary Grant (Roman Holiday) to Kurosawa (The Hidden Fortress).
In part two, I’ll focus on how leanback programming can be implemented to create the Me TV experience.
Editor's Note: This is a vendor-written case study. StreamingMedia.com accepts vendor-written case studies based upon their usefulness to our readers. Albert Lai is CTO of broadcast and media at Brightcove.
For streaming video providers, the choice now is to either build in-house or to look to the cloud and get out of the technology business.
Delivering custom video streams to every user demands vast changes in how publishers operate and how they think about content.