Walk into any commercial real-estate office and you'll find a map with a circle drawn on it. The circle is in red marker, or it's a tidy GIS overlay, or it lives in a deck slide labeled "trade area." In every case, the circle is wrong. Not approximately right. Wrong.
The reason is geometric. A circle measures distance as the crow flies. Customers don't fly. They drive, they walk, they take transit; and the thing that determines whether they show up at your store is not how many miles you are from them — it's how long it takes them to get to you. Time is the unit. Distance is a proxy that fails the moment the network does anything interesting.
The math, briefly
A circle of radius r covers an area of πr². At r = 5 miles, that's 78.5 square miles — about 50,000 acres. Inside that circle, you'll find some land you can reach quickly (a straight road that points away from your store) and some land you can't reach at all (the other side of a river without a bridge). The circle treats them identically.
An isochrone — a polygon of equal travel time — gets this right by construction. It's drawn by simulating travel away from the origin in every direction along the actual road network, stopping at the chosen time. The shape that results follows the roads, bends around water, and reaches further along corridors that move quickly.
DEFINITION
Isochrone. A line connecting points reachable in equal time from a fixed origin. The interior of the line is the service area at that travel time.
Where the disagreement is biggest
We ran the comparison across 412 U.S. retail-relevant origins — strip centers, downtown blocks, and shopping-mall anchors. For each, we computed the population inside a 5-mile radius and the population inside a 15-minute drive at 5:30 PM local. The two numbers agreed within ±5% on only 11% of origins. Disagreement above 25% was the median.
DATASET · 412 ORIGINS · MAY 2026
Radius vs. drive-time disagreement
The cases at the extreme were predictable in retrospect. They lived near coastlines, rivers, mountain ridges, and highway interchanges — any geography that bends the network away from a perfect grid. They also clustered in cities with strong polycentric structure (Minneapolis–St. Paul, Dallas–Fort Worth) where the radius assumption of a single origin's gravity falls apart.
Three failure modes
The disagreement isn't random. It tends to fall into three buckets, each of which produces a different type of bad decision.
1. The water failure
Any origin near a coastline, lake, river, or bay. The radius circle sweeps over water and counts demand that simply isn't there. Worst case: a beachfront store whose radius extends a mile into the ocean. Coastal cities lose 10–30% of "addressable" demand the moment you switch to drive-time.
2. The highway failure
Origins near a freeway interchange can over-reach versus their radius — a 15-minute drive may extend 6 or 7 miles in one direction, well past the 5-mile circle. Operators that use circles will underestimate competitive overlap with the next store down the interstate.
3. The grid failure
Dense one-way grids (parts of Manhattan, San Francisco, downtown Portland) compress drive-time polygons unpredictably. A 5-minute drive in midtown Manhattan covers less than a square mile; the radius analogue would suggest ten times that.
What to use instead
Use the polygon. For most retail decisions, a 10- or 15-minute drive isochrone is a better trade-area definition than any circle, full stop. For walkable retail, use a 10-minute walk. For franchise territory carving, use the polygon to define the territory rather than defending a meaningless circle in a contract.
And don't stop at one polygon. Run the same origin at 5, 10, and 15 minutes; the ratios between them tell you something about the local network's character. A site whose 10-minute polygon is dramatically bigger than its 5-minute one sits near a highway access point — capacity to draw from far away, but vulnerable to a competitor on the same highway.
"The circle was a tool from when the alternative was a paper road atlas and a ruler. Both of those are gone. There's no reason left to keep the circle."
The counterargument, briefly
Circles are simple. A circle can be drawn by anyone, on any map, with no API call. For a back-of-envelope screen of fifty potential sites, a circle is genuinely faster, and the loss of precision is acceptable because the precision wasn't needed yet.
The honest place to use a circle is the very first pass. The first thing you should do once you've shortlisted is replace every circle with a polygon. DriveZone's free tool is sufficient for that work, and you can compare radius and drive-time side by side on any origin.
Summary
- Circles measure distance; isochrones measure time. Customers care about time.
- The disagreement is large: typical median above 25%, sometimes above 40%.
- Water, highways, and dense grids are the three failure modes.
- Use the polygon for any decision past the first screen.
- For franchise territory carving, draft contracts in polygon terms — they'll hold up better.
∎ END OF ARTICLE