Once a chain has two nearby stores, some revenue from the first will shift to the second. That shift is cannibalization, and it is nearly unavoidable when expansion fills an existing market. The issue is how large the shift becomes and whether the combined economics of two stores improve on the economics of the single store that came before.
That calculation is often incomplete before a lease is signed. Teams emphasize the proposed site's volume while missing revenue that will move from existing locations. A new store may reach its own target as a nearby established store weakens beyond the expansion model's forecast. Network revenue can therefore stay flat or fall after adding a unit.
Why Hidden Costs Grow
Revenue moving from an established store to a new one is cannibalization's clearest cost and belongs in any credible projection. Other costs are harder to see and can build on that transfer. Because they are less predictable, they are often left out of the pro forma.
Fixed cost absorption is one example. An established store that loses substantial volume to a nearby opening still carries the same occupancy costs against less revenue. If its margin depended on a certain volume, the lost revenue hits the bottom line almost entirely until costs come down, a process that often takes longer than a quarterly cycle. A store moving from 100 to 80 units of revenue with 70 units of fixed cost is structurally different from one at 100 with 70 in fixed costs.
Brand investment dilution is another. Market marketing and awareness spending once supported one store. After a nearby opening, that same investment must produce returns across two locations. The new store also receives some attribution from the brand presence the first store built, creating shared overhead that is almost never modeled directly.
Staff retention adds a third cost. Experienced employees at the established store, especially those with local customer relationships, may move to the new site by choice or at the company's request. The original store then bears training and familiarity costs that take time to restore.
When Cannibalization Still Makes Sense
We are not arguing against second or third locations in an existing market. Cannibalization can make economic sense when market demand is sufficient for two stores to cover more than one, when the new site wins demand previously served by a competitor, or when greater network density adds convenience and raises overall market share.
Consider a simplified hypothetical: one store in a mid-size metro captures an estimated 60 percent of addressable demand within a 12-minute drive. The other 40 percent is either going to a competitor or unmet because the drive is too long for some customer groups. A second site covering that underserved area could win much of the remaining demand while taking only some visits from the first store. If network revenue after cannibalization exceeds the prior total, and the new site's incremental contribution exceeds its total cost, the expansion works economically despite the transfer.
The comparison is the net contribution of two stores against the one store baseline, not the proposed store's standalone forecast. That requires a direct estimate of cannibalization at the first store rather than dismissing it because it is difficult to model.
Estimating the Transfer
A sound estimate starts with where existing customers come from and how many would find the new site meaningfully more convenient. Foot traffic data with estimated home origin zones for visitors to the existing store provides the raw input. Customers whose homes are closer to the new site are more likely to shift visits after it opens.
A gravity model formalizes the estimate by assigning each candidate location a patronage probability based on distance and store attractiveness. Its output distributes expected customers across the two store network, yielding total network volume and an existing store cannibalization rate. For this use, gravity models improve on radius methods because they reflect road network geometry, which shapes real convenience more than straight line distance.
This analysis does not produce a precise forecast. It provides a range calibrated to the existing site's geography and customer origin patterns. Using the midpoint as the pro forma base case and the upper end as a stress test gives the lease decision firmer grounding than ignoring cannibalization or relying on a guessed percentage or rule of thumb.
Cannibalization Across the Network
Over time, separate cannibalization decisions form a network pattern. Chains that routinely understate the effect can build portfolios in which many stores modestly miss their original pro formas. The cause becomes hard to identify later because each store was gradually affected as the network expanded around it.
A stronger practice is to make cannibalization modeling part of every new site review, not a special exercise only when proximity looks troubling. Establishing that discipline at 20 to 40 locations, before expansion speed makes it unwieldy, supports better network economics than ad hoc, inconsistent assessments.