Any geographer knows points, lines, and polygons form the basis of vector data analysis. These three data types enable data scientists to effectively map, visualize, and model physical attributes on Earth’s surface. When it comes to physical places, polygon data is essential for getting a true representation of a property or building. While point data for places has its use for competitive analysis, creating trade areas, and other key business intelligence operations, there are many mission-critical use cases that require building footprints or geometry.
A polygon is defined by Esri as “a GIS object that stores its geographic representation—a series of x and y coordinate pairs that enclose an area—as one of its properties (or fields) in the row in the database.” While there are many ways to represent a point - such as through building centroid, street-level geocode, or address geocode - polygons by definition must represent the physical boundaries of a property or building. There is no “close enough” for polygons. If it’s inaccurate, it’s wrong.
The main reason why precision and accuracy are so fundamental to polygons is that the most common use cases for geometry or building footprints require a truthful representation.
This one may be obvious, but should not be overlooked. While all geospatial data is intended to be mapped in some capacity, mapping polygons almost always means creating an accurate visualization of a property or building’s boundaries. When creating a map, if the decision is made to use polygons over points to represent places, there is most likely a desire to represent how properties relate to one another in the most detail possible.
Mapping building footprints or geometry reveals important relationships between physical places. With detailed spatial hierarchy information, polygon data can be mapped to show which places are located within another place. Accurate and precise geometry also enables visualization of co-tenancy or adjacency, which can inform site selection or risk assessment strategies. Mapping polygons also provides information related to the accessibility of a place, such as where an entrance is located or where the closest parking lot is for that location.
Polygon data can also be more informative of consumer interaction with a place than point data alone. This is particularly true for advertisers or retailers deriving store visit attribution. Mobility data makes it possible to see how consumers move throughout the day, but without contextual location information, that movement is meaningless. When combined with polygons, mobility data shows which places people visit and how long they stay there.
But as with mapping, this information is only valuable if it is accurate and precise. Accurate polygons are critical for correctly attributing visits to specific places. Visit attribution that relies on a centroid radius is likely to both under- and over-attribute visits to places, particularly when places are located close together or within the same structure. Spatial hierarchy and accurate polygons ensure visits are correctly attributed to places, which boosts the efficacy of location-based marketing and retail analytics.
Polygon data’s ability to showcase co-tenancy and adjacency with precision is fundamental to insurance risk assessment. Measuring and modeling property risk is dependent on what is going on at or near a specific location. For example, a nail salon that opens in a strip mall next door to a fireworks store will have a higher risk profile than one that opens across the street, or next to a grocery store.
In an increasingly competitive and data-driven insurance market, insurers must stay on the cutting edge of risk assessment and modeling to win and retain customers. There is no room for error, or data that is “good enough.” Under-assessing a property’s risk can lead to increased exposure for the insurer, while over-assessing can lead to customer churn and dissatisfaction. Polygons give insurers the precision they need to ingest into their models to assess risk with confidence.
Our sole focus is places data. From understanding where POIs are located, to what their boundaries look like and how they relate to places around them, and even how consumers interact with them, SafeGraph is the expert in physical places. This maniacal focus enables us to curate and deliver polygons of the highest accuracy and precision to power mapping, visit attribution, risk assessment, and more.
SafeGraph Geometry data provides building footprints or geometry for POIs in the US, Canada, and UK, along with essential metadata. We include spatial hierarchy information to help users understand the relationship between places within the same polygon, as well as brand and parking lot details. Our metadata also includes a field that indicates whether the polygon was generated from machine learning or hand-drawn, providing full transparency into our data curation process.