Hitchhiker's Guide to Streaming Media: Online Video Advertising
>> Lookup> Advertising Models
It has been widely speculated that the online advertising spend in 2008 will surpass the TV advertising spend. The jury is still out to see if these forecasts will bear fruit, but what is clear is that, for the online video user, advertising is becoming more prominent and prolific.
There are so many models already in place and many have subtle (and less subtle) differences, so for the sake of this guide entry lets group them into themes:
Banner adverts or “floating” Flash adverts that are actually part of the webpage that is also providing access to the stream. Revenue works in exactly the same way as banner adverts.
Short video adverts played before the requested video stream begins. Revenue works in CPM like banner adverts, but usually attains 3x to 5x the revenue seen from banners since the user is deemed to have focused themselves on the video window and is waiting through the ad for the content he or she actually requested. One of the most successful models in the streaming media space.
Short video adverts played at regular intervals through the delivery of a live or long-form stream. These are very similar to traditional TV ad models. Revenue is derived again from a CPM but will be higher again than Pre-Roll and considerably higher than Banner models since the user is deemed to be “locked in” to a stream and will put up with viewing through adverts to continue watching their content. A new and emerging model since it relies on live or long-form content and these are newer to mass market than short clip offerings, which are largely dominated by pre-roll solutions.
New, sexy, and still emerging Overlay adverts provide some form of animated textual or graphic content which is displayed as a layer overlaying the video. Some of these can be pre-programmed to work with and around the video content, while others will deliver ads with no regard to the video. The upside is that they don’t generally interrupt the underlying video, so the user continues a relatively uninterrupted experience. However they do distract from the video and are unpopular in many circles.
• Sponsor Package:
Advertisers have two choices for packages. They can often opt to be a major sponsor of an online video. This may mean that their advert is ALWAYS played to every video user. This may be considerably more expensive, and often the deal will not be CPM-based but rather based on a time window (for example for an “event” or for “a month”) and the rights will be sold as a single flat fee package. For close association with entertainment and sports events this type of marketing can be a very good way for financing an event webcast, while providing maximum exposure for the brand.
• Agency Package:
The second package option would typically be provided by an Advertising Agency. Here the Agency procures ad space on a wide number of video streaming sources or services and offers their advertiser a certain number of plays of their advert across all these (or perhaps a subset of these) sources. This is less targeted but has wider reach than a single sponsorship package may. These packages are generally the backbone of CPM models in the video pre-roll (and increasingly interstitial) advertising models.
• Collection/Data Capture:
A less common system used is “forced collection” where users can only access a stream once they provide a valid email address and consent (opt out options are often present but obscure) to receive postings from the service providers sponsor(s). This email address is then used for targeted mail shots by advertisers who essentially pay for the list.
There are doubtless others, too, but these topics broadly cover the different ways that advertising packages are often thought of when advertisers look at placement options and the potential returns.
There are several parties who need to come together to make a system like this effective. The first is the content provider, who generally dictates who the ad providers are by selling an ad package. These are then taken up either directly by sponsors, or by an agency. The content provider will generally contract a production and distribution service provider and instruct them to play the sponsors or ad-agencies pre-roll(s) out in front of each stream or at intervals therein. In the sponsorship model often the video is provided and streamed from the same platform and by the same service providers who are delivering the actual content. In the Agency Package model the distribution system will usually be configured to include an advert, which may or may not be hosted and distributed by their own systems, and then the user subsequently connects to the requested content. This means that the distribution provider does not have
to be involved in the content management of the adverts.
In either model the agencies have a number of systems that track advert plays independently of the content distribution system, and it is their own figures that they ad agency will calculate revenues from. There are even a few ratings systems emerging that are attempting to provide this audit service as an “objective third party” for both the advertisers and distribution service providers.
The author has also seen one or two interesting technologies emerging from this industry. There is a system that can detect ad-breaks in terrestrial transmissions and insert alternative, internet-specific advertising solutions on the fly as the stream is encoded, and this system can go further by actually forcing the end users players to pull different adverts for different users (so called “user targeted” advertising) and all these methods are driving a premium on the cost to the advertiser.
The real value of the internet is proving very alluring for advertisers. Unlike in TV models where advertising was largely untargeted and “eyeballs” hard to count, the internet is providing the same advertisers with more than just exposure. The “connectedness” of the Internet is also allowing ad-campaigns to also become market research tools since a great deal more is known about users and their usage patterns.
It hardly surprising that it is rapidly emerging as a multi-billion dollar industry.