SINGAPORE: A number of factors are to blame for unseen mobile ads, but slow loading times are the biggest problem when it comes to mobile viewability, senior agency executives have said.
Most consumers are now mobile first in Asia, while, globally, 64% of all online videos are set to be viewed on mobile devices by next year, according to Zenith's Online Video Forecasts 2016 report.
Viewability concerns are growing for advertisers. While ad fraud and ad blocking are on the rise, slow loading times are the number one reason behind why consumers don't see mobile video ads: 62% of users stop watching videos if the load time is very high; 53% of mobile site visitors abandon the video when it takes longer than three seconds to load.
"The biggest challenge that we see today is that ads are loading really slow compared to the content itself," said Vinay Uttamchandani, InMobi's Director of Products - Brand & Programmatic, in a recent webinar hosted by Mobile Marketing Association APAC.
He put the turn-off time at even less than three seconds. "If it (the video) doesn't load within two seconds, it's not viable. Users just drop off," he reported. (For more, read WARC's exclusive report: Understanding mobile video viewability for brands.
Ad clutter from publishers is a particular bugbear. "There are too many trackers on ad servers … slowing down the process for the ad itself to load," he said.
Another complaint is the visual clutter offered up by publishers. "There are too many ads packed in by the publisher on the screen which makes it extremely difficult for any of the advertisers to make their brand propositions aware to end users, " Uttamchandani said.
The solution? Move through the native ad experience by using the technology of the connected devices, he argued.
"In a true native in-app environment … pre-cached and buffer-free experience can result in a much more better performance. That is what we see the industry moving towards," he said.
"You don't have to choose performance over individual measurability. You can get both for the same method."
Data sourced from WARC