ellipsis flag icon-blogicon-check icon-comments icon-email icon-error icon-facebook icon-follow-comment icon-googleicon-hamburger icon-imedia-blog icon-imediaicon-instagramicon-left-arrow icon-linked-in icon-linked icon-linkedin icon-multi-page-view icon-person icon-print icon-right-arrow icon-save icon-searchicon-share-arrow icon-single-page-view icon-tag icon-twitter icon-unfollow icon-upload icon-valid icon-video-play icon-views icon-website icon-youtubelogo-imedia-white logo-imedia logo-mediaWhite review-star thumbs_down thumbs_up

The art of in-app video ads (part 2 of 5): Conquering latency

The art of in-app video ads (part 2 of 5): Conquering latency Daniel Kang
Mobile video ads are one of the most powerful ways to engage users. The average click-through-rate of mobile video campaigns increased by 265.7 percent in 2013 and the share of tablet and mobile video plays soared by 74 percent. Mobile video ad spend is expected to increase by 82.1 percent in 2014.

Video watchers have been trained by TV to expect a seamless transition from content to ads. A large body of research shows that users have no tolerance for low quality videos. University of Massachusetts Professor Ramesh Sitaraman published a study titled “Video Stream Quality Impacts Viewer Behavior” where he measured the correlation between online video load delays and user drop-off rates. Based on data representing 23 million video views from 6.7 million unique visitors, he found that 20 percent of viewers give up and leave after just two seconds. Every second of additional delay results in approximately 6 percent more viewers jumping ship. Four out of five online users will click away if a video stalls while loading.

Speed is critical, and latency is one of the biggest hurdles preventing programmatic mobile video ads from reaching their full potential.

What causes latency?

Latency is the period of delay, usually measured in milliseconds, required for a video to load. A lot happens within those milliseconds. A request gets sent to an ad server, which tries to find the best ad to serve to that user. As a result, one or multiple ad providers may be called to make this determination.

Once the right provider has been found, the creative is downloaded to the app from a CDN. The user sees a video ad, but those millisecond hops between servers and mobile content delivery networks (CDNs) add up.

Keep in mind that this scenario assumes that everything works perfectly. There are multiple points of failure that contribute to latency, including inefficient ad server infrastructure, CDN location, and interactive ad units with huge file sizes. Latency can also be caused by poor bandwidth, poorly-written SDKs, and legacy device hardware. The last is a particularly relevant concern on mobile.

Overcoming latency

Solving the inherent latency issue with mobile requires anticipating and optimizing for the ad experience. This can be achieved in a couple of ways:


The first is by pre-caching the video payload well ahead of the ad event.

Pre-caching allows video ads to load and buffer in the background while the user is engaging with the app’s content. When the time comes for the ad to appear, it will play instantly, regardless of the connection speed and status before or during the experience.

Not having the video buffer, lag and judder enables users to register the core message of the advertisement without distractions.  This can make all the difference in advertiser ROI, whether the video is a performance ad that drives users down a funnel or brand ad that helps grow recognition and awareness.

Hybrid serving

A second, newer technique is hybrid serving. Most ad tech companies either stream their videos to mobile devices or pre-cache them. Both methods have their advantages and disadvantages.

Streaming ads are delivered in real time to the user, which means only the most current and relevant ads are shown. Streaming is ideal for time-sensitive campaigns, such as hyperlocal promotions, which must be targeted to users in a particular place at a particular time.

In addition, digital marketers have become accustomed to buying streaming media over the last decade. Staying within the same buying model makes buying mobile ads easy.

The downside of streaming is that ads may buffer, even on fast connections, which ruins the user experience. Pre-cached ads circumvent this shortcoming entirely, but don’t work for all types of advertisements.

Dynamically selecting whether to stream or pre-cache based on conditions such as connection type, device make/model and ad campaign goals will give you the right foundation to serve ads intelligently. This means serving ads that users actually like, instead of just serving the ad with the highest CPM. The result is higher ROI for advertisers and more revenue to publishers.

A blended approach to serving guarantees that ads delivered are both relevant and engaging, with the lowest amount latency possible.  It’s a win-win for users, publishers and advertisers.

The need for speed

Getting people to pay attention to ads on mobile can be challenging enough under the best of circumstances. Video ads have the unique ability to attract and engage users in a way static images can’t. Creating a positive user experience is crucial, so video ads need to be consciously presented in just the right way, at just the right moment.

As technology evolves, connection and processing speeds will get faster. The challenge for now is seeking elegant solutions that get latency as close as possible to zero.

Daniel Kang is the product lead at Vungle.  He is currently in charge of the Vungle direct platform and has built out Vungle's Programmatic Exchange, the worlds first mobile video focused exchange.  Prior to Vungle, Daniel worked at Google...

View full biography


to leave comments.