Integral Ad Science , a global media measurement and optimisation platform, announced a made for advertising (MFA) AI-driven site detection and avoidance product. The company’s MFA site technology aims to improve transparency into advertiser campaign quality, identify where spend is being allocated, and inform optimisations to minimise waste on MFA sites.
MFA sites are web pages featuring low quality content (such as spam sites, or ad farms) created solely to serve ads, and are optimised to perform well against traditional verification metrics, such as viewability. However, advertising spend on these sites does not drive meaningful outcomes, such as conversions or brand lift.
IAS’s new product leverages AI to uncover MFA sites at scale, allowing advertisers to take back control of their media quality and cut down on wasted spend. During Alpha testing, IAS delivered comprehensive campaign analysis demonstrating superior MFA site identification for some of the world’s largest advertisers and agencies.
IAS’s product supports the Association of National Advertisers’ (ANA) recent definition of MFA sites and incorporates characteristics such as ad-to-content ratio, ad refresh rate, and the source of the traffic coming to the site to classify a site as MFA. According to the ANA’s Programmatic Media Supply Chain Transparency Study, 21% of all advertisement impressions measured were served on MFA sites.
“Our MFA product was built to deliver unprecedented transparency to advertisers and provide them with the ability to both detect and avoid MFA sites in order to redirect their ad spend to publishers that drive a return,”Yannis Dosios, chief commercial officer, Integral Ad Science, said.
Advertisers and their agencies need confidence that the industry is converging on agreement about the specific websites that compose the MFA category. By training its model against Jounce Media’s widely adopted list of MFA domains and incorporating signals from Sincera, IAS has developed the industry’s first pressure tested solution for detecting and blocking MFA at scale.
The IAS MFA product completed alpha testing in early October 2023 and is available now as a beta measurement offering for select customers. General availability is expected to expand to all customers in early 2024.