A new type of report from Pixalate finds that incomplete or inaccurate supply-path reporting is associated with significantly elevated invalid traffic (IVT) rates in programmatic OpenRTB buys.
IVT, which is traffic generated by bots or other non-human means, includes ad fraud as well as unintentionally invalid traffic.
The new analysis focuses on supply chain objects (SCOs) — an identification system introduced in 2019 by IAB Tech Lab to allow buyers to see all parties involved in selling or reselling a given bid request.
In Q1 of this year, Pixalate found that 87% of SCOs were marked as “complete” within the SCO system, and that paths marked as “incomplete” exhibited 55% higher IVT rates.
In Q2, a second analysis dug deeper, including a focus on SCOs that are marked as complete and therefore supposedly valid.
Pixalate used an SCO validation process that evaluates whether a supply path legitimately originated from a publisher and whether that publisher has a direct relationship with the first seller in the supply chain, and also measured IVT rates across the analyzed supply paths.
The company monitored more than 20 billion
CTV/OTT, mobile in-app and programmatic ad impressions during Q2 and compiled validation data during the month of June.
The results: While 72% of measured traffic with SCOs marked as complete passed Pixalate’s validation process, 19% of these failed validation.
Further, the average IVT rate among the 19% of “complete” SCOs that failed validation was 65% higher than the rate for traffic marked complete that passed the validation process (chart top of page).
In Q2, IVT rates for SCOs marked as incomplete
were found to be 41% higher than the rate for traffic marked complete that passed validation — a small improvement over Q1.
The analysis also studied the results of redundant supply paths.
Nearly three quarters (71%) of traffic with SCOs was found to have the single, unique path prescribed by best practices, but 8% of traffic had three or more paths.
Redundant supply paths resulted in higher IVT rates even among SCOs that were marked complete and validated by Pixalate.
Compared to traffic with a single path, traffic with two paths had a 17% higher IVT rate, and traffic with 11 or more paths had a 24% higher IVT rate.
Reason: more paths mean less direct control over where inventory passes through, leading to more opportunities for IVT.
While stressing that the SCO and other IAB Tech Lab transparency initiatives have given the industry the tools to understand the industry’s complex advertising supply chains, the report concludes that additional validation steps are required to ensure that the “complete” flag in the SCO is accurate.
The overall results support a policy at Basis Technologies that bans sellers believed to be intentionally misrepresenting or withholding supply chain information, or repeatedly failing to maintain accurate records, says the company’s platform operations manager, Ayse Pamuk.
“Our approach is based on the hypothesis that lack of transparency only benefits the bad actors,” she says. “The findings in Pixalate’s report indicating elevated IVT rates on supply paths that are either marked incomplete or inaccurately represented as complete confirms our hypothesis.”