Geofencing Best Practices: Polygons vs. Radius

Geofencing sounds simple on paper: draw a virtual boundary around a location and capture any device that enters it. In practice, the shape and precision of that boundary has a massive impact on the quality of your data and the accuracy of any analysis built on top of it. Radius-based fences are easy to produce at scale but structurally lossy; custom polygons are labor-intensive to build but produce visit data that holds up to scrutiny. GSDSI's POI & Geofencing product delivers polygon-based boundaries, and the MRC measurement standards codify why the distinction matters for attribution.

Key Takeaways

  • Radius-based geofences default to noise — circles don't match the shape of real-world buildings.
  • Polygon geofences reduce false-positive visit counts by roughly 25–45% in dense urban and mixed-use areas.
  • Radius still makes sense for large outdoor venues, parks, and awareness-level campaigns where reach matters more than precision.
  • Your vendor's polygon construction and refresh cadence are the core diagnostics — ask how boundaries are built, validated, and kept current.

How Radius and Polygon Geofences Actually Work

Radius fences take a latitude/longitude center point and a distance (typically 50–200 meters) and capture every device event that falls inside the circle. They are trivial to generate programmatically, which is why most self-serve targeting platforms default to them. Polygon fences trace the actual footprint of a building or parcel — corners, setbacks, parking lots, loading docks. They require parcel data, satellite validation, or manual drawing. GSDSI's POI & Geofencing product is polygon-based and refreshed against parcel and satellite sources on a recurring cadence.

The 25–45% False-Positive Gap in Dense Areas

The gap matters because circles don't match reality. A 100-meter radius around a coffee shop in a dense urban block might capture three other businesses, a sidewalk, and a bus stop — every device that pauses at that bus stop gets counted as a "visit." Internal GSDSI testing and public-domain methodology work consistently find that switching from radius to polygon attribution reduces false-positive visit counts by 25–45% depending on surrounding density. In shopping malls, transit hubs, and mixed-use developments the improvement is even more dramatic. For the broader quality framework, see the geospatial data quality framework.

When Radius Geofencing Still Makes Sense

Radius isn't always wrong — it's the right tool for a narrower set of jobs:

How High-Quality Polygons Are Actually Constructed

A defensible polygon is built from multiple inputs: parcel data (where available), satellite imagery validation, and operator-level corrections. The polygon should cover the indoor footprint of the business, not the full parcel — parking lots and loading docks belong to different attribution rules. Refresh cadence matters because retail locations open, close, expand, and remodel constantly. For the POI-quality angle, see why POI data quality makes or breaks foot-traffic analytics. Census and NAICS-aligned classification — reference Census Business Dynamics Statistics and the NAICS 2022 manual — are what let polygon-tagged visits roll up correctly by industry.

What to Ask a Geofencing Vendor Before You Sign

Use these diagnostics when evaluating a provider:

  1. How are polygon boundaries constructed — parcel data, satellite, manual drawing, or a blend? What's the QA process?
  2. What's the refresh cadence, and how are openings, closings, and remodels reflected?
  3. What's the published false-positive rate, and how was it measured? (Compare against MRC measurement standards and IAB quality guidelines.)
  4. For chain locations, how is the polygon differentiated from adjacent same-chain locations in dense clusters?
  5. How are indoor vs. exterior visits separated? (A device that walks past a storefront isn't the same as a device that enters it.)

Answers to these questions tell you more about the downstream quality of any analytics built on that foundation than any top-line coverage statistic.

Frequently Asked Questions

How much more accurate are polygon geofences than radius geofences?
In dense urban and mixed-use areas, polygon geofencing reduces false-positive visit counts by approximately 25–45% compared to radius-based attribution. In shopping malls, transit hubs, and tightly-clustered retail, the improvement can be larger. In large outdoor or low-density contexts the gap narrows because radius circles don't overlap as many adjacent locations.
When should I still use radius-based geofencing?
Radius works for large outdoor venues, parks, and event spaces with irregular natural boundaries; awareness-level campaigns where reach matters more than attribution precision; early-stage market research; and any case where a trusted polygon simply doesn't exist — a well-scoped radius with a confidence caveat beats a polygon you can't verify.
How are high-quality POI polygons constructed?
Defensible polygons blend parcel data, satellite imagery validation, and operator-level correction. The boundary should trace the indoor footprint of the business, not the full parcel — parking lots belong to different attribution rules. Refresh cadence is critical because retail locations open, close, and remodel continuously. GSDSI's POI & Geofencing product follows this methodology.
What questions should I ask a geofencing vendor before signing?
Ask how polygons are constructed, what QA process is applied, what refresh cadence keeps them current, what published false-positive rate they measure against, how clustered same-chain locations are differentiated, and how indoor vs. exterior passes are separated. The geospatial data quality framework walks through the full diligence checklist.