Video ad tech provider Cedato has introduced its Contextual Lookalike Targeting technology, which uses machine learning to analyze performance data from billions of videos ads in order to decide when and where to serve a new ad to suit an advertiser’s KPIs. The new technology leverages Cedato’s Predictive Knowledge Graph, which is based on data from 400 billion plus video ads.

Cedato is positioning its lookalike technology as an alternative to user-based targeting, which has been curtailed due to cookie limitations by Apple/iOS and the implementation of GDPR. Cedato said the technology is GDPR-compliant and privacy-friendly. Enhanced targeting is one of the main benefits of programmatic advertising, which Cedato specializes in.

Cedato highlighted a hypothetical auto campaign that was seeking to replicate KPIs accomplished on a prior campaign. The new lookalike technology would identify the previously used inventory and analyze the campaign’s parameters. Then, using its Predictive Knowledge Graph, Cedato would find the optimal inventory for the new campaign.

The new lookalike technology is the latest innovation from Cedato. Earlier this month the company introduced its “IntentView” video ad format which features an ad playing as a picture-in-picture window for 5-6 seconds with the audio off.