FAQ
Common questions about Geosocial Proximity sources, time windows, volume, and how it differs from demographic systems.
What are the data sources?
Most of the data are geotagged social media posts (primarily Twitter/X and Instagram). See Methodology.
What is the time window?
Scores use a running 24-month (two-year) window. That span accounts for seasonality and tends to be predictive for retail use cases. Vintage labeling is described in Methodology.
How is this different from MOSAIC, PRIZM, and similar systems?
Geodemographics such as MOSAIC and PRIZM classify areas from the demographics of residents, then often infer behavior from surveys. Proximity scores areas from what people are doing and saying on social media in that place. Two demographically similar areas can still differ in social behavior in ways that affect sales and models.
Is social media mainly used by youth?
No. Social platforms include substantial older usage, and Proximity segments can skew younger or older than the national mean depending on the segment. Treat age as one of many attributes in segment metadata—not as a reason to dismiss the dataset.
How is volume accounted for?
A volume index proportional to post activity in an area supports weighted averages for trade areas. See Working With the Data and Dataset Structure.
Do posts come from where the behavior was tagged or the user’s home block group?
Posts are attributed to where the behavior was geotagged. Tighter trade areas usually better reflect what is happening near a store or site.
Related
- Methodology
- Working With the Data
- PersonaLive FAQ — household segmentation questions