What a ring assumes
A radius assumes travel is equally easy in every direction at every speed. It ignores highways, rivers, bridges, rail lines, one-way systems, mountains and the fact that a customer on the far side of a freeway with no crossing is not your customer at all.
What changes when you route it
Routed catchments are irregular. They stretch along arterials and stop dead at barriers. Two stores the same distance apart can have completely different overlap depending on whether a road connects them.
The effect compounds at scale. Across a national footprint, ring-based counts systematically overstate reach in dense grids and understate it along highway corridors.
Why most tools still use rings
Routing is expensive. Computing every store's catchment on a real road network means solving millions of shortest-path problems. The usual workaround is to approximate with a ring, or to buy a device panel and infer catchment from observed visits — which substitutes one approximation for another.
Narrative Geo does not sample. Every trade area is routed on the actual road network with OSRM, against full Census/ACS population and a maintained POI set. The drive-time matrix is precomputed at 1.08 billion rows, so a national query across a 20,000-store chain returns in seconds rather than being metered per location.