Foot traffic data has become one of the most widely used inputs in retail site selection over the past several years. The aggregation of mobile device location data at a commercial scale has made it possible to see, with reasonable accuracy, how many devices visit a specific location, where those visitors come from, and how their visit patterns compare to nearby competitors. That information is genuinely useful. It is also genuinely incomplete as a basis for a location decision.
The problem is not that foot traffic data is inaccurate. For measuring visitation volume and basic catchment patterns, it is usually reliable enough to work with. The problem is that foot traffic tells you about current visitor behavior at existing locations. It does not, by itself, tell you how much demand for your category exists in a trade area, how much of that demand is already being captured by existing supply, or whether a new location would generate genuinely new visits or primarily redirect visits from a location you already operate.
What Foot Traffic Measures
When you look at foot traffic data for a potential site, what you are typically seeing is one of two things. Either you are looking at traffic to an existing retail tenant at that location, which tells you something about that specific tenant's current performance but may not generalize to a different concept. Or you are looking at general area traffic, which tells you how many devices pass through or stop in the general area around the site.
Area traffic is a reasonable proxy for the potential customer pool that physically passes near the location. A high-traffic corridor has more potential customers flowing through it than a low-traffic one. But the conversion from potential customers to actual store visits depends on factors that foot traffic data does not measure: whether those passing devices belong to people who are your target customer, whether they are in a buying mode for your category, and whether they have satisfying alternatives already available nearby.
A site in a high-traffic retail corridor surrounded by well-established direct competitors serving the same customer need is a very different opportunity than a site with the same foot traffic volume in a market where your category is underrepresented. Both show the same number on the foot traffic report. The actual demand opportunity is substantially different.
The Demand Side That Foot Traffic Misses
Local demand for a retail category is not visible in foot traffic data. It requires a separate analysis of what consumers in the trade area actually spend in your category and how much of that spending is currently being captured by existing supply versus going unmet or being fulfilled elsewhere.
This demand analysis draws on consumer spending data at the trade area level, which shows category-level spending totals derived from aggregated transaction records. Comparing that spending total to the estimated category revenue capacity of the existing supply in the trade area gives you a rough demand gap estimate: how much category spending exists in the market above and beyond what current retailers are capturing.
A site with high foot traffic in a market with a positive demand gap is a much stronger opportunity than a site with the same foot traffic in a saturated market. The foot traffic number looks identical on both evaluations. The demand situation is fundamentally different, and that difference will show up in performance after opening in ways that the foot traffic analysis did not predict.
The Cannibalization Question
High foot traffic at a candidate site is a positive signal. But if your chain already operates locations nearby, some fraction of the foot traffic at the new site will come from customers who currently visit your existing stores. That is cannibalization, and it is invisible in raw foot traffic counts.
The relevant metric for a location decision is not total foot traffic at the candidate site but net new visits: visits from customers who are not already being served by your existing network. Estimating that requires knowing the origin zones of the people likely to visit the candidate site and comparing them to the origin zones of your existing stores' customer bases. The overlap between those two populations is your cannibalization exposure.
A site with high total foot traffic and heavy origin zone overlap with a nearby existing location you operate may produce less net new revenue than a site with somewhat lower foot traffic but a catchment that is genuinely distinct from your existing coverage. Foot traffic analysis without cannibalization analysis systematically overestimates the opportunity of in-fill sites relative to sites that extend your coverage into new territory.
Visitor Quality Signals Within Foot Traffic Data
This is not to say that foot traffic data is limited only to volume counts. More sophisticated use of foot traffic data can provide quality signals that go beyond aggregate visitor totals. Visitor home origin distribution tells you where customers are coming from and gives you a rough trade area map. Visit frequency, measuring how often the same devices return within a 90-day period, signals customer loyalty and engagement levels. Dwell time analysis can distinguish between quick transactional visits and longer engagement visits, which varies by retail category in informative ways. Cross-shopping patterns, showing which other retailers customers visit in the same trip, reveal the co-tenancy relationships that drive visit co-occurrence.
These signals add context that raw visit counts do not. But even this richer set of foot traffic metrics does not answer the demand gap or cannibalization questions directly. Those require external category spending data and your own network's origin zone analysis, which are separate analytical inputs.
The Complete Picture
A site selection decision grounded in foot traffic data alone will systematically overweight certain signals and miss others. It will do reasonably well at identifying high-activity corridors and established retail zones. It will do poorly at distinguishing between markets where category demand is genuinely unmet and markets where the demand has already been captured by existing supply. It will produce no information about cannibalization risk from the new site's relationship to your existing network.
Foot traffic is the starting point, not the conclusion. The additional analysis, specifically demand gap estimation and cannibalization modeling, is not optional complexity for large chains with sophisticated teams. It is the part of the evaluation that explains most of the performance variance between locations that looked similar at the foot traffic level but diverged significantly in actual results.
A well-scored location analysis that combines foot traffic, demand gap, and cannibalization data narrows the range of outcomes significantly. It does not eliminate them. Site execution, local marketing, competitive response, and macroeconomic conditions all contribute to outcomes in ways that no pre-opening analysis fully captures. But the analytical foundation matters, and foot traffic alone does not provide it.