Recently, I came across a programmatic video ad that used a nested player to override the playback controls of the actual video player in the page, meaning the only way to mute or pause the player was to close the browser completely.
This is a frustration to both the end user and the site owner, and the fact that a hack like this made its way onto a reputable network is a huge problem in ad quality and indicative of even bigger issues in our industry that continue to get buried.
Digging further, I found there were a total of 11 impression trackers where I’d normally expect only two. These types of layered tags negatively impact viewability and lead to problems like degraded user experience, latency, or no ads playing at all. What’s more, this particular ad facilitated false completion data for the video campaign.
So how did we get here?
The bits that break
Advertisers and agencies have increasing demands to report on viewability and sometimes deploy rogue, questionable hacks to get around them. Client side ad insertion, for example, may facilitate a form of analytics measurement, but it also facilitates the possibility of ad fraud that breaches ad quality, performance and yield.
So is bad ad management like this a product of laziness, ignorance, aptitude or negligence?
In a normal workflow between ad-ops and ad sales, there are little or no verifications in place to check supplied third-party creative for retargeting or unwarranted user tracking that is likely to compromise most publishers’ data policies. This problem extends to viewability measurement methods that have been implemented at the request of an advertiser or agency.
On a technical level, this would be fine if it did not also impede the amount of time a video ad has to load. After all, the more you load onto a truck the longer it takes to get up to the desired speed. If a video ad takes too long to resolve, it will be killed off in the request mechanism by the ad server.
The implications of having poorly developed, untested, or rogue code in an ad placement is bad news for any kind of ad quality, server side or client side. Unfortunately, the filter is at the end of the funnel and it’s the broadcasters and publishers that wind up bearing the brunt of poor ad quality, which, in turn, affects viewers.
We, as an industry, need to ask some tough questions: Is it a ploy to obtain better completion stats to satisfy a commercial arrangement with a media agency? Do audio controls get overridden to meet audibility goals unbeknownst to a publisher, ad-ops team, or supply-side platform (SSP)? Does the tracking implemented on the inventory breach data policy?
Getting on the server side
Hidden hacks are leading to missed opportunities to monetise inventory with programmatic and are letting critical viewability and data policies be compromised.
By now, server side ad insertion (SSAI) should not be a new concept. The mechanics of it have been floating around for quite a while in the digital landscape and it’s a good solution to the monetisation question many publishers are facing. There are a number of benefits to SSAI, the best of which are related to how it can improve user experience while meeting better viewability requirements, not to mention that it’s a proven method for circumventing ad blockers.
There are some who haven't fully come around to SSAI as a solution, arguing that it doesn’t always work. But when there’s a loophole in the network that allows for unverified creative to run, is it really SSAI that is broken? Or is it the mechanism that allows for inventory like the ad I recently discovered to be delivered?
Cleaner ads, cleaner delivery
Clean ads lead to cleaner delivery which actually means better measurement and monetisation. The point here isn’t to point fingers or nitpick at the various methods used in video advertising, but to improve quality across the board.
At the end of the day, a focus on ad quality reduces error rates and increases yield; the idea is to capitalise on controlling the inventory publishers and agencies manage. We’re all working towards making things better for end users — this should be done with the understanding that when publishers and agencies create an environment for video ads to serve successfully, yield actually improves.
It’s about customer-centricity. Simple as that.
Justin van Emmerik is a technical specialist at Brightcove ANZ