Part 1 in our series of key issues for building your video tech-stack
In our programmatic video series we compiled for you a host of setup and run issues to consider when planning and assembling a video ad-tech stack.
Our goal is to navigate you through the process, simplify it and provide you with best practices. We’ll begin the series with general considerations then dive into the nitty gritty.
Selecting the right video ad-tech stack is tricky. With so many devices, formats and fragmented data in the market place, making the right choices can be overwhelming. It’s difficult enough to build a technology stack for display operations. In programmatic video the complexities grow and the expertise needed reaches a new level.
Getting the combination wrong can severely impact your business results, so you want to carefully look at your technology, structure and video technology partners before taking the plunge.
Starting from square 1:
What do I want to accomplish with my video tech stack?
On a strategic operational level you want to focus on how to drive your cost /performance ratios. To streamline performance, you’ll need to consider tools that deliver simplicity, scale and efficiency. On a functional level you’ll need solutions that act on the right audience at the right time and provide a seamless cross-platform experience. When transacting in programmatic video, publishers in particular need real time access to maximum demand through video-specific bidding technologies.
Since programmatic video has rapidly evolved into a full scale process, new video-first tools are vital to assemble the layers in the stack and optimize the workflow. Re-purposing a display stack or combining legacy systems with new technologies seems a prudent choice at first but it could do more harm than good. It could generate cross-screen, format and loading time issues that lead to an increase in risk, reduced revenue and a compromised viewer experience.
What structural issues to consider?
When structuring a programmatic video tech-stack or improving an existing one, each new product adds a level of complexity, opening up process and workflow issues.
Player interfaces, bidding technologies and optimization layers need to be assembled with minimal complications and fewer point solutions. A scalable structure is needed to support the various forms of programmatic trading; RTB, Direct deals, etc., and manage tens of thousands of transactions per second from multiple demand sources.
Video workflows need to operate automatically and seamlessly together to closely integrate the different layers and third party sources, maximize yield and improve viewer experience. Publishers looking to optimize their fill rate in real time need to consider a yield optimization layer with algorithms and analytics for extending video audience reach across all screens, and delivering lift.
The case of the smart light player
While there is no shortage of video players not all are created equal. In reality different players serve various types of video ads in different video formats.
VAST (Video Ad Serving Template) provides a framework for embedding video ads, tracking them across all screens and devices and enabling efficient and effective monetization. Simply put, it’s a piece of XML script that’s supposed to tell the player which video ad to play and how long it should show. It also provides important tracking information such as impressions, completions etc., VAST 4.0, the latest IAB standard update, could eliminate the need for VPAID by taking over much of its interactive role. The catch is that many players still don’t support the newer VAST versions.
Today many publishers and marketers prefer a player that meets their technical and business needs without sending their team into video ‘Admageddon’. Simply put they favor a lighter player with minimal coding customization and plugins.
The efficiency of a player highly depends on your ability to play valid VPAID and VAST ad tags for web, mobile and apps and simply monetize. In theory, you can run any third party video player ad with a VAST tag because the video player is supposed to understand and display it.
The case for a bidding technology designed specifically for video
Programmatic video stacks today offer varied forms of bidding technologies. Many are based on display processes that won’t consider the complexities of playing video. Till today some still rely on the drawn out waterfall bidding process. Others offer single RTB bidding or Server side only header bidding solutions.
All these technologies have a major drawback; they fail to verify and authenticate video ads, largely affecting the yield, fill rate and viewer experience. Such drawbacks are particularly widespread when floor prices are high, as in the case of video.
To run video advertising effectively, VPAID and VAST formats need to be authenticated on the client-side of the video player. But if triggered on the client side alone, the process can create repetitive ad requests that slow loading time. Ideally, a video stack should comprise a bidding technology that can perform the predictive work on the server side, while completing essential ‘last mile’ VPAID authentication and verification on the client side.
A holistic programmatic video tech-stack needs to address multiple challenges. It needs to meet diverse format demands, cross-screen platforms and fragmented environments as well as a host of other complex issues specific to video.
In part one of our series we touched on a few important considerations. Stay tuned for the rest of our series as we look more closely under the hood.