Trade Area & Drive-Time Analysis for Retail & Franchise

Ask most retail expansion decks how big a store's catchment is and you will get a number attached to a circle: "a five-mile radius" or "a ten-minute drive". The circle is easy to draw and easy to defend in a meeting. It is also, almost always, wrong. Real customers do not travel as the crow flies. They travel on roads — around rivers, over bridges, through congestion, along one-way systems and motorway junctions. A defensible trade area is therefore not a radius at all. It is a drive-time, walk-time or transit-time isochrone computed on the actual network.

This distinction is not academic. The shape of the catchment is the input to every other number in a site decision: how many target customers a location can reach, how much of their spend it can realistically capture, and how much demand it will quietly steal from stores you already operate nearby. Get the shape wrong and every downstream estimate inherits the error.

Why a radius is a lie

A circle assumes the world is a flat, frictionless plane with roads running in every direction at equal speed. No real geography behaves that way. Consider two sites with an identical "10-minute" claim:

• A store beside a motorway junction may reach twenty kilometres of dense suburb in ten minutes along the carriageway, while reaching barely two kilometres on the congested side streets behind it. Its true isochrone is a long, lobed shape stretched along the road corridors — nothing like a circle.
• A store on the far bank of a river with only one bridge for miles has a catchment that is amputated on one side. Half its "radius" is water and severed street grid it can never reach in ten minutes.

Model both as circles and they look interchangeable. Model them as network isochrones and one is clearly a stronger site than the other. The moment you switch from Euclidean distance to network travel time, the map starts telling the truth.

From catchment shape to demand

An isochrone on its own is just a boundary. It becomes a business case when you intersect it with the population and spending inside it. Overlay the catchment on gridded population and demographic data and you can size the addressable market directly: how many households, in which income and life-stage segments, fall within a genuine ten-minute drive — not within a circle that happens to include a lake and a golf course.

This is where trade-area work connects to the broader discipline of spatial analysis for market decisions. The catchment is the spatial unit; demographics, segmentation and spend data are the attributes you attach to it. Together they turn a shape on a map into an estimate of reachable demand.

Cannibalisation: the number nobody wants to see

If you already operate in a market, the most important — and most frequently ignored — question about a new site is how much of its demand it will pull from your existing stores rather than from competitors. Two catchments that overlap are two stores fishing in the same water. Model that overlap before you sign a lease and you can quantify cannibalisation as a percentage of transferred demand, not discover it a year later in a same-store sales report.

Practically, this means computing the isochrones for the proposed site and every nearby incumbent, measuring where they intersect, and allocating the contested population between them. A site that adds coverage in a genuine gap is worth far more than one that simply relocates sales from three streets away.

Gravity and Huff models: who actually goes where

Overlap tells you catchments compete; it does not tell you how customers split between them. For that, retail geography uses spatial interaction models — the gravity model and its retail refinement, the Huff model. The intuition is simple: a shopper is drawn to a store in proportion to its attractiveness (size, assortment, brand) and inversely to the effort of reaching it (travel time). Given a set of competing stores, a Huff model estimates the probability that a household at any location patronises each one, and from there each store's expected market share.

This is what lets you move from "these two catchments overlap by roughly 40%" to a concrete split — this new store captures a fifth of the contested demand, the incumbent keeps most of the rest, and the remainder leaks to competitors. Those probabilities are the difference between a site decision and a site guess.

Network analysis for coverage gaps and territory design

The same network engine that draws a single isochrone scales up to two portfolio-level questions:

• Where should the next store go? By computing coverage across the whole road network, you can find the populated areas that no existing or competitor location reaches within an acceptable travel time — the genuine white space — and rank candidate sites by the net new demand they add rather than the demand they cannibalise.
• How should territories be balanced? The same routing graph lets you design and balance sales or franchise territories so each is compact, contiguous and carries a comparable workload or market potential — a classic districting problem that only makes sense on network distance, never on straight lines.

Choosing between candidate sites once demand, cannibalisation and coverage are quantified is itself a ranking problem, which is why serious site selection so often ends in a multi-criteria parcel screening that weighs travel-time reach against rent, competition, footfall and brand fit.

An interactive way to see it

The fastest way to make the "radius is a lie" argument land is to show it. Cartolytic builds interactive isochrone tools that draw the real drive-time catchment directly beside the naïve five-mile circle, so the difference is visible at a glance rather than argued in a footnote. A companion overlap view takes two proposed sites, renders both true catchments, and flags the contested area between them — turning cannibalisation from an abstract worry into a shaded region on the map you can measure. These are the kinds of decision tools we deliver for retail, franchise and multi-unit expansion teams.

Data sources

Robust trade-area analysis blends open network and demographic data with commercial routing, foot-traffic and segmentation data used to calibrate the results.

Open & public

• OpenStreetMap road network (routing via OSRM, Valhalla or GraphHopper)
• GTFS transit feeds for public-transport catchments
• US Census / ACS, Eurostat and national statistics offices
• WorldPop and JRC GHSL gridded population
• Overture Maps

Commercial

• HERE, TomTom, Google and Mapbox routing & isochrone APIs
• Esri Business Analyst (built-in drive-time plus demographics)
• Placer.ai and SafeGraph / Advan (real foot-traffic to calibrate catchments)
• Experian Mosaic and Nielsen / Claritas (household segmentation)
• CoStar (retail real-estate data)

Tools

• OSRM, Valhalla and GraphHopper for network routing and isochrone generation
• Gravity and Huff spatial-interaction models for patronage and market share

From map to lease decision

Done well, trade-area and drive-time analysis produces a compact set of outputs a real-estate committee can act on: catchment maps that show true reach, demand estimates sized on real demographics, cannibalisation percentages between proposed and existing sites, and a ranked list of site recommendations. Every one of those depends on getting the catchment shape right first. If you are scoping an expansion programme, this is exactly the kind of question we frame at the start of a GIS-driven site screening — before the first site is drawn on the map.