Technology company Criteo has released an updated version of their prediction engine, able to process 15m predictions a second and accurate to within a 1 in 10,000 chance of occurrence. Following the engines release, the company is seeing a 38 per cent increase in sales while maintaining the same ROI.

The update has been three years in development and predicts both the chance of a user clicking on an ad, and their likelihood of making a purchase on the advertisers site before deciding whether to buy ad impressions. This refines an adverts reach to the greatest number of potential buyers within the defined cost-per-sale target.

“The Criteo Engine is now processing huge volumes of data with improved models, and as a result we are able to predict evens that happen only once out of every 10,000 times – like when an ad impression generates a click and a sale,” said Jonathan Wolf, chief product officer at Criteo. “As a result, we help advertisers make the best possible decisions that will maximise the reach and effectiveness of their marketing.

“This is where advertising needs to be and we are very excited to continue to prove the value that can be generated from focusing on pure post-click performance.”