Trade area overlap vs. cannibalization: what's the difference?

Trade area overlap is geometry. You draw a catchment around an existing store and another around a proposed site, usually as drive-time polygons or radii, and overlap is the share where the two shapes cross. It is a property of the map.

Cannibalization is what happens to the business: the portion of the new store's sales that comes out of the existing store rather than from new customers or from competitors. One is measured in shared area, the other in moved demand. A reporting tool can hand you an overlap figure in seconds, and that figure is where the analysis starts.

1. Area overlap is the raw geometry, the percent of the two catchments that intersect. It treats every acre the same, whether it holds ten thousand households or none.
2. Demand-weighted overlap takes that same intersection and weights it by what sits inside it: population, households, daytime workers, or category spend. A narrow sliver over a busy downtown can outweigh a wide band over farmland.
3. Demand transfer estimates the sales that actually move, using a gravity or Huff-style model that accounts for distance and how much pull each store has. This is the cannibalization number, and it is almost always smaller than the area overlap that produced it.

In Geod those last two rungs are reported separately, a demand-weighted overlap figure and an estimated transfer, alongside the net-new and competitor-capture portions of the forecast. A site brief never passes off raw overlap as the cannibalization risk.

Because the first rung ignores demand, a small overlap and a large one can swap places once you weight them. Picture a candidate whose catchment clips a corner of an existing store's territory, but that corner is the dense, high-spending core where most of the existing customers live. A larger overlap spread across thin suburban edges can move far fewer dollars.

So an overlap percentage on its own is a weak proxy for risk. The same number can hide very different amounts of demand. Independent analyses of real catchments show this gap routinely, with a small fraction of overlapping area accounting for a much larger fraction of a store’s customers.

An overlap figure stays silent on the things that decide a deal. It does not say how much demand sits inside the shared zone, or how loyal those customers are to the store they already use. It says nothing about whether the new site is more or less attractive than the existing one, which is what determines who switches.

It also ignores the other side of the ledger: the net-new demand the candidate reaches beyond the overlap, and the share of its forecast it wins from competitors rather than from your own store. Two sites with identical overlap can have very different net effects once those factors are counted.

• Capacity relief: when an existing store runs at or near capacity, a nearby unit absorbs demand the busy store was turning away through long waits or limited hours, which recovers sales rather than only shifting them.
• Delivery and pickup density: tighter store spacing shortens delivery routes and improves coverage for online and pickup orders, which can raise total orders even as the two locations share dine-in customers.
• Competitive defense: a unit planted inside your own overlap keeps a competitor from taking that ground, so a measure of transferred demand becomes the price of holding the position.

The test in each case stays the same. The new unit has to clear your return hurdle once the recovered and net-new demand are weighed against what it transfers.