Unlock innovation with the most accurate dataset of US commercial POIs and their visitors.
Thank you! Your submission has been received!
Oops! Something went wrong while submitting the form.
Industry leading location stacks run on SafeGraph
Business listing and building footprint dataset for places people spend money (~5MM POIs in US).
Place traffic and demographic aggregations that answer: how often people visit, where they came from, where else they go, and more.
Seamlessly integrate your existing POI data with SafeGraph's enriched Places data.
Truth seeking requires complex data organization
All of the signal, none of the noise
Complete POI Information
Business category, open hours, visit count, popular times and more are associated with each POI. The top 3,000 brands are mapped to over 1MM POI and noisy locations are removed (ATMs, kiosks, etc.).
detailed & accurate geometry
Building geometry is derived from satellite imagery and places are spatially hierarchical. Polygons for sub-stores are specified and tied to their parent store geometry (e.g. Starbucks inside a strip mall).
Algorithmically combining the best of all data sources
Onboard data from several diverse sources. Compare, de-dupe, cross-reference, and discard bad data.
Complex polygons are drawn by hand and others are derived from satellite imagery.
Use machine learning and human feedback to programmatically merge POI metadata and geometry.
Algorithmically classify brands and spatial hierarchy. POIs (like restaurants) can exist within other POIs (like airports).
SafeGraph leverages unique truth sets to continually improve the accuracy of our Places.
Places updates are delivered on a monthly basis.
SafeGraph is just a data company
That's it – that's all we do. We seek to understand the physical world and power innovation through open access to geospatial data. We believe data should be an open platform, not a trade secret. Information should not be hoarded so that only a few can innovate.
Ready to analyze or develop?
Get in touch to learn more or receive a data sample.