• by Eric Hoffert , March 3, 2016

Consumer broadband speeds are in the tens or hundreds of Megabits, there are literally hundreds of millions of individual video creators around the world, and TV viewership is declining in favor of digital devices.

But as advertising dollars begin to follow consumers’ eyeballs, there is a serious problem the industry must solve in order for digital video to reach its full potential: latency. We’re all frustratingly familiar with the spinning buffering icon, and it is time to do something about it.

Today, video advertising vendors run sequential video waterfalls or conduct mediation in parallel. In many cases, the video player waits until the slowest video demand source returns a result, which can cause major delays.

 

A vendor will often try to sell a video impression for a publisher, fail to do so, and never pay a penalty, all while the viewer waits and watches a black screen and spinning cursor. Not only does this no-penalty arbitrage open the door to inefficiency and latency, but it also takes an enormous bite out of publishers’ advertising dollars. We can do better.

Here are five steps that can be used to get there:

1. Move latency away from the video play button and onto servers

It isn’t necessary to wait until a user has pushed play and is waiting for their video content to begin running a client-side auction. In fact, there are already a couple of better alternatives. One option is to leverage thousands of low-latency real-time bidding (RTB) servers to run auctions instead of consuming the CPU on personal computers.

 

Similarly, server-side ad stitching also moves the auction process away from the client by allowing publishers to stitch their video and ad content together at the CMS level rather than the browser level. As a bonus, this technique also thwarts ad blockers.

2. Retrieve video instantly with video ad caches

 

Once an auction or video mediation process has run, the winning VAST XML can be kept in a video ad cache. This stores the VAST result, not the actual video ad, and so it is relatively lightweight from a storage and bandwidth perspective. HTML5 local storage can be used so that video ad XML is immediately available when it’s time to play.

3.Rewrite VPAID mediation creatives so auctions start VPAID InitAd and run quickly

The VPAID standard was invented by the IAB as a valuable way to add interactivity to video advertising content. But vendors often use VPAID in different ways, such as client-side auctions. Some run these client-side auctions when the video ad should be starting to play (VPAID StartAd), which can cause many seconds of blank loading time.

These vendors should change their code so auctions run upon initializing (VPAID InitAd) to separate the time of auction from when the video creative is supposed to begin playing. In addition, it’s time to make VPAID InitAd much faster when running client-side auctions.

4. Shift video demand from VPAID Flash to VPAID JavaScript or VAST with video files

Flash consumes a tremendous amount of resources—as much as 80% of CPU and a potential doubling of memory utilization. To prepare for the inevitable Flashpocalypse, advertisers and agencies need to train their artists and developers to transition to open Web standards (e.g. HTML5) and mobile friendly environments (hello JavaScript!).

This approach also supports video viewability verification. Another fast option is to use VAST files with direct links to video creatives for direct video ad serving.

5. Develop new IAB standards for client-side mediation

Latency is an industry-wide issue that cannot be solved unless it is addressed more broadly. We need standardized interfaces so auctions are transparent and can be controlled, whether on the client or the server.

Efforts such as the IAB LEANAds (Light, Encrypted, Ad Choice, Non-Invasive supported) program have started to promote user experience as the rise of ad blocking poses a threat to the open Internet, and similar actions specifically designed for video need to be taken.

The VAST 4.0 standard holds promise to reduce latency on a number of fronts, including server side stitching, conditional ads, and the separation of viewability, but many question if this goes far enough. It is worth noting that prebid.is an open source industry project that reduces display advertising latency and increases monetization is producing positive results for publishers.

Similarly, Accelerated Mobile Pages is an open source project aiming to speed mobile browsing. There is yet to be an open source project that focuses on fixing video latency, this could be an opportunity for the industry to work together.

 

 

With a better, faster video Internet, we all win.

 

Advertisers can reach relevant, engaged audiences. Users enjoy a fluid, TV-like experience. And publishers, particularly the independent content creators who provide a diversity of rich, alternative content to the walled video gardens of Facebook and YouTube, can grow and maintain their fan bases while winning advertisers’ growing digital video ad budgets.

 

This is a call to arms to change the way video monetization works, and time is of the essence.