With the exponentially increasing use of video ads, many do not stop to think about the downsides they bring with them as well. While I don’t believe that the challenges of latency, data drain, malvertising, or autoplay sound warrant not serving video ads, I think it of utmost importance to know your defense tactics for handling them.
In the article, QA in the Videoscape: Keeping Quality in Line with Demand, Brian LaRue skillfully delves into the challenges video ads present and discusses the solutions.
Excerpted from the article:
Sure, increased video consumption has been great for monetization, from the perspective of many publishers. But rapid development often produces a lot of mistakes and oversights, and it certainly has in digital video. Latency, data drain, malware--these are all factors that contribute to bad ad experiences (and worse), and they can bring the revenue train to a crashing halt. The ever-shifting variables involved in quality assurance only get more complicated as video consumption on mobile devices continues to grow. But these issues are not forces of nature, and publishers who are aware of them are exercising a variety of solutions, workarounds and preventative measures to assure quality video.
Tired of Waiting for Video
If we were all okay with waiting and waiting after hitting “play” on a video, we might not have given away our VCRs. Once users experience immediacy, it’s hard for them to turn back. On the open web, if a video doesn’t load quickly, that’s sufficient cause for the user to navigate away and toward something faster post haste. So when the cause of that latency is a third-party ad that won’t load--that load time needs to get sorted out.
The causes for latency are, unfortunately, myriad and diverse. Among publishers we spoke with, many, but not all, of the causes they cited fell under the general umbrella of vendor or player issues. Multiple vendors handling the same execution can be a technical bear, and it
can produce pesky discrepancies.
One specific cause of latency cited is the necessity that DSPs and SSPs talk to each other efficiently, and the fact that they don’t always do so. For example, the buy side can sometimes fail to deliver an ad in a timely fashion, or neglect to time itself out if it passes on an impression. These kinds of snafus can come about when an SSP hands off control to a DSP or other buy-side partner, and then the DSP doesn’t hand control back to the SSP. When that happens, it’s not the DSP or the SSP that times out, but the ad server itself, which might take 10 or so seconds.
Other latency culprits might include the player itself, VPAID errors (in which, when VPAID is used as a buy-side decisioning engine, the winner throws its own nested auction and, if there’s no winning bid, returns and empty VAST wrapper), poorly-encoded or even non-encoded videos, issues with vendor content delivery networks, scripts from third-party vendors (verification, viewability, etc.), or ad serving being handled several links in the chain away from the publisher.
Clearly, rooting out and solving for the core latency problem or problems is a chore unto itself. But publishers have found it can be managed and even improved, in spite of how much troubleshooting and how many intermediaries might be involved. Publishers can ask buy- and sell-side vendors to implement their own time-out functions, beyond the ad server’s universal time-out. It’s recommended that ops know how to read VAST, and also check tags in browsers that either take or block cookies.
Publishers should work with buyers and intermediary vendors to troubleshoot, because the core problem in latency might be right under their noses. One publisher told us that after thoroughly checking internally and with DFP, they determined the latency was coming from an online video player, which was adding seconds of rendering time before the ad was served.
The number of factors at play calls for a continued dialogue between internal teams on the publisher side and between publisher and vendors.
Other publishers are using vendor tools, like GeoEdge or developing their own verifications, to root out errors and their likely causes. Completion or error rates can reveal possible problems without getting too far in the weeds. And it should go without saying, but publishers ought to vet vendors thoroughly and seek out feedback from other publishers about those vendors’ strengths and weaknesses.
To read the full article QA in the Videoscape: Keeping Quality in Line with Demand, please click here.
Just like Brian says, “Poor user experience can lead to declining publisher revenues, but with a little foresight, preparation and prevention, your audience can end up satisfied and none the wiser about what it took to get there.” So be in touch with us at GeoEdge, and we’ll help you keep your users safe and engaged.