Discussions

In the documentation of methodology differences between Advan and SafeGraph Patterns, section "No Modeling of Visits/Visitors in Shared Geometries", it describes how a traffic drop was over-stated for NAICS codes including restaurants due to a traffic drop in parks that shared polygons with establishments in these NAICS codes. It says "Advan is actively working on a long-term fix and is running further analyses to understand mitigation techniques and further causes." Has a long-term fix been implemented?

In our research, we are looking at total restaurant visits in Chicago using the Monthly Patterns Data, and we indeed see this drop in visits in November and December 2022. However, we see a spike back up to trend in March 2023. Is this because Advan resolved the issue? What is the reliability of the data now? What are recommended adjustments to make, if we would like to accurately estimate total visits to establishments in a given NAICS code over time? (I read all of the documentation but did not find a satisfying answer for how to adjust for issues like this.)