Panelists opt-in by installing a mobile app that measures location in the background, persistently and discreetly. The data measured includes source, latitude, longitude, accuracy, accelerometer and orientation. In addition panelists provide demographics and respond to survey questions. This robust set of data is the foundation for Placed Insights.
INFERENCE MODELING & NORMALIZATION
Location and metadata are run through the Inference Pipeline to assign a place to a series of raw data points without the use of check-ins or journals. The pipeline is continually validated and trained against a growing set of reference data. This data is then normalized using industry best practices combined with proprietary models to reflect the behaviors of the U.S. population.
The output of the Inference Pipeline is the set of Placed Insights reports, reflecting the intersection of the digital and physical worlds, quantified through the Place Graph. With Placed, businesses are now able to leverage innovative analytics previously limited to online measurement, to gain real world insights on the paths and places that people visit.