Not All Dealership Visits Have Equal Value

In the automotive industry, it is often difficult for companies who pay for advertising to get lower funnel metrics based on consumer behavior in the physical world. In response, advertisers use offline attribution to leverage offline visitation as a lower-funnel performance metric for analyzing and optimizing automotive campaigns.

As a result of working closely with automotive advertisers on offline attribution, Placed identified a need to categorize visits by visit intent. This white paper details our process to develop and validate a machine learning model that provides visibility into dealership visit intent for use in Placed Attribution reports.

Reasearch Methodology:

The research was based on more than 75,000 completed surveys to Placed's first-party audience. The survey's collection began in June 2017 and continues to be active today. For each completed survey, we joined the survey responses with the visit cluster that led to the survey push and then also with the demographic attributes of the user that generated the visit. This series of joins enabled us to extract all the necessary features to build an accurate predictive model.