Getting to Connected TV Ubiquity Bit by Byte (Part 2)

Albert Lai's picture
Albert Lai on February 15, 2013

In Part 1 of this post, I focused on the intertwined issues of audience and content programming. In today's post, I will discuss the technical issues surrounding Connected TV device platform capabilities--or, in many instances, limitations--as well as the "softer" but equally important issue of promotion.

Device Platform Capabilities

Welcome to one of the more headache-inducing areas to consider, as each device platform (and, in some cases, the model) has its own set of technical capabilities and limitations when it comes to video experiences.


In most instances, the publisher will build an application, or some set of services that interacts with a proprietary service, native to the device platform. The publisher should understand the development and on-boarding process from skills and timeline perspectives.

  • Skills: What is the development language (C#, JavaScript, Flash)? Are the required skill sets available in-house? If not, is this expertise worth building via internal training and/or hiring - or buying through contractors or partners/agencies?
  • Timeline: What is the process for launch? Some device platforms require weeks to months for certification.
  • What is the process for ongoing releases? Some device platforms have only scheduled windows when updates are allowed.
  • Deployment: What is the process for establishing a global footprint? The ability to manage video experiences on a geographic basis is relevant for publishers that have content-licensing restrictions or want to implement a smarter content programming strategy.


In the Connected TV ecosystem, one of the most commonly overlooked areas is metadata management. The user interaction with a Connected TV requires consideration of both input device--typically limited actions via a remote control or, in less common cases, a difficult-to-use keyboard, motion or voice activation--and screen real estate for a lean-back experience. This typically means several changes that may require a significant amount of re-thinking:

  1. Defining a content hierarchy or content structure that can be optimized for a lean-back experience with limited user input gestures
  2. Modifying content metadata to be more easily and consistently rendered without common desktop-oriented user interface motifs (e.g., scrollbars, hovers, drag-and-drop)
  3. Ensuring images are optimized for the larger screen size, e.g., increasing quality using lossless formats or increasing the default quality settings of lossy formats, increasing the resolution of images for use in a 1920x1080 environment, preferred aspect ratio (e.g., organizing content for an overall 16:9 device, use of square content layouts in the Xbox or use of grids in Sony BIVL)

Content Encoding, Delivery, and Protection

Each device platform has its own supported set of capabilities for playback of content. Publishers will need to determine whether the video content codecs and containers (e.g., MP4, HLS, HLS with AES-128, Smooth), content delivery (e.g., RTMP, RTMPE, HTTP, HTTPS), and content protection (e.g., Widevine, PlayReady) capabilities are sufficient - or whether the built-in capabilities can be enhanced through third-party or custom development. The variety of formats and device platforms exposes another potential cost: the additional work required to transcode, store, and deliver video renditions of additional formats and likely higher bitrates as device platforms encourage and require high-definition content (typically 8-10Mbps renditions to support 1080p).


Each publisher has to determine the business value of the Connected TV application. The hard dollar costs to develop, launch, and operate this digital initiative often must generate some form of physical dollars. Each device platform should be evaluated by its ability to support the monetization strategy, as the device platform may restrict monetization to a particular ad server (a third-party or proprietary system). And, the publisher must often negotiate business terms in the form of revenue share or a similar model.

Publishers may choose to leverage display advertising or in-stream advertising, but they must also understand the limitations of these formats in a lean-back experience. Both display and in-stream formats are one-way; the notion of a click-through is uncommon and likely impractical. Non-linear overlay advertising during video playback, while becoming more popular in desktop environments (YouTube, ahem), is a much more challenging format within context of a lean-back experience, as the user has limited ability to interact with the ad unit. And, in implementations I've witnessed, the ad unit often feels unnatural, covering lower-thirds and open and closed captions. Advertising creative, whether in-stream or display, should also be formatted for a 1080p experience.


Publishers will be responsible for measurement, from measuring user behavior across sessions to video playback and video quality of service. Publishers may utilize any number of analytic services (e.g., Google Analytics, Omniture, comScore, Nielsen, etc.) but the question I urge publishers to answer is, "How are users interacting with content in context of the device platform?" In these lean-back experiences, the remote is the enemy and quality is king.

The upside in the lean-back experience is that the intent of the user is to explore, choose, and consume content. The website notion of driving more page views per session - directing the user to click-click-click through navigational gyrations - is not an effective approach. Publishers should be thinking about trading clicks for engagement, for instance, reducing the navigation friction to increase the amount of time a user is watching the content and, ostensibly, increase the monetization opportunities. Mechanisms should be added to understand how content programming is affecting content selection. Is specific content better than others? If so, is it due to the duration, genre, promotion, or position? Most importantly, these numbers should be put in context of the consumption experience, i.e., understanding content performance on the Connected TV platform vs. mobile vs. desktop. Publishers will likely see differing performance from one Connected TV platform to the next, and understanding which content is "working" will be core to the success of the aforementioned content programming strategy.

A quick aside: Ever wonder why some films are displayed with alternate names in cable VOD? A content distributor once whispered to me their frustration. Not only did the older systems restrict title length, but users were less apt to scroll-scroll-scroll to the bottom of the list. Consequently, the distributor used metadata positioning (i.e., adjusting the name whenever possible) to change how a user could discover their content, ultimately affecting content performance.

Previously, I mentioned the issue of encoding for the device platform, often resulting in the need to deliver higher quality video renditions than for other form factors (i.e., desktop, smartphone, etc.). Publishers should ensure instrumentation is in place to provide telemetry data to understand the video quality of service, including which renditions their users are viewing and to what extent users are experiencing buffering during playback. User expectations are high, as users are comparing the Connected TV video experience not to the desktop or mobile experience but to the broadcast television model. With quality-of-service data, publishers can make more informed choices about their encoding profiles. For example, information about the average available bandwidth and the average rendition bitrate can determine whether the encoding profiles should be adjusted. If content is encoded for adaptive bitrate streaming at 1.0Mbps, 2.5Mbps, and 5.0Mbps but the average available bandwidth is 2.1Mbps, the publisher should reassess their encoding profiles to optimize the rendition profiles based on the user or device platform capabilities.


We can't cut down a tree without the right tools, but if a tree falls in a forest … well, that's how promotion fits into the equation.

  1. The most straightforward approach for the publisher is direct promotion to its existing desktop and mobile audience. The overall hope is that by increasing the availability--ubiquity--of the publisher's content, the audience will not shift their consumption but rather augment their consumption through the additional channel.
  2. Publishers should ensure that the device platform has an explicit plan to help build awareness of the offering. These methods could range from "in-store" digital promotion to email marketing to offline marketing initiatives (print, television, events). Without this collaborative marketing, the publisher may lose the opportunity to build awareness outside of their core audience.

Eeny, Meeny, Miny …

Very few publishers have the content or capability to create and maintain the ubiquity that Netflix has established. Publishers that understand the limitations--both the fragmentation and their own capabilities--can succeed if they are able to select and optimize for Connected TV ecosystems that enable them to cost-efficiently make the upfront and ongoing commitments to delight their users.

As NAB approaches, the anticipation grows as we wait for announcements of an evolving Connected TV future, one that is likely even more fragmented but also filled with technology partners that are working to help publishers navigate this road.

Until then, I'm in the mood for a walk to another dimension with Rod Serling--now where is that little red button?

Post new comment

The content of this field is kept private and will not be shown publicly.