On Device Research launches fraud prevention system

The system, initially trialled for three months, has proven to increase the quality of respondent data by 30% in the UK and 40% in the US compared to the programmatic survey market average.

Behaviour Trap programmatically scores respondents’ quality by ingesting different indicators that influence the quality of a respondent, such as biometric data captured from the mobile device, app usage behaviour, demographic profile information and negative reconciliations from survey exchanges. The data is then processed and machine learning is applied to give respondents a quality score in real time.

The live system automatically catches fraudulent respondents, limits the participation of overactive survey participants and has the ability to respond to the evolving threat of bad actors.

Available through On Device Research’s Curious Cat app, Behaviour Trap ensures data quality for both its advertising effectiveness surveys and programmatic survey responses. The company estimates that in excess of three million surveys will have been completed by the end of 2021 and anticipates more than five million will be undertaken in 2022.

Alistair Hill, chief executive and co-founder of On Device Research, commented: “Launching our Behaviour Trap system allows us to stay ahead of the evolving threat of fraudulent survey respondents and ensure our clients make decisions based on the best quality data possible. The sophisticated and evolving nature of the survey means that a sophisticated and evolving system is required to combat fraud.

“By using the mobile devices sensors and an adaptive system that understands behaviour in real time, we can ensure quality survey data is collected on an on-going basis.”