The volume of invalid traffic that has been detected on desktop browsers has fallen by more than one third this year, according to the second wave of the IAB/PwC Invalid Traffic Benchmarks.
The latest data, which PwC assesses from Comscore, IAS and Moat data shows a decrease in invalid traffic from 3.7% of impressions between July and September 2016 to 2.4% between October 2016 and March this year.
For mobile, the drop was even more stark – from 3.8% down to 1.3%. Only 1.4% of video inventory was deemed fraudulent, the first time monitored for the first time.
Ad fraud is believed to cost the Australian market about $116 million each year, with $68 million wasted on fraudulent website traffic, such as bots.
The IVT benchmark tracks one element that could indicate ad fraud, the amount of internet traffic that is invalid not seen by a human.
The traffic can include general invalid traffic such as bots, spiders, crawlers and other routine means of filtration, as well as sophisticated invalid traffic such as hijacked devices, ad tags or creative, malware and misappropriated content.
The lowdown on IVT detection
There are two levels of fraud detection used by third-party verification firms – general invalid traffic (GIVT) and sophisticated invalid traffic (SIVT).
According to the IAB, GIVT includes traffic identified through routine and list-based means of filtration, such as bots, spiders, other crawlers; non-browser user agent headers; and pre-fetch or browser pre-rendered traffic.
SIVT includes traffic identified through advanced analytics, multipoint corroboration and human intervention, such as hijacked devices, ad tags, or creative; adware; malware; misappropriated content.
Comscore and Integral Ad Science both provide SIVT-accredited detection while Moat offers a GIVT solution.
IAB Australia adheres to the MRC's global recommendations, which state: “All vendors are required to comply with the General IVT (GIVT) provisions in the guidelines, but vendors are strongly encouraged to adopt Sophisticated IVT (SIVT) provisions.
“Vendors adopting only GIVT provisions, by definition will be applying incomplete filtration (exclusive of SIVT) and as a result, will provide incomplete assurance that traffic is human.”
GIVT detection can often pick up higher levels of invalid traffic because it considers all data centre traffic as invalid, even though research shows data centre IVT is closer to 29%.
This is because corporate firewalls are often connected to data centres and can link to thousands of IPs used by office workers and other legitimate sources.
IAB Australia director of research Gai Le Roy tells AdNews GIVT detection should be followed with further measures that provide a granular assessment of data centre invalid traffic.
“In the IVT Guidelines, vendors are required to empirically support identification of invalid traffic or filters as well as to analyse and minimise material false positives resulting from them,” she says.
“If publishers are being asked to block or not pay for inventory from certain IP addresses due diligence must be done to ensure legitimate audiences are not being blocked.”
Have something to say on this? Share your views in the comments section below. Or if you have a news story or tip-off, drop us a line at adnews@yaffa.com.au
Sign up to the AdNews newsletter, like us on Facebook or follow us on Twitter for breaking stories and campaigns throughout the day.