Retail foot traffic data measures how many people visit a physical location, how long they stay, how often they return, and where they travel from. Retailers use it two ways: to measure marketing and store performance after the fact, and to judge a site before signing a lease on it.
What it measures:
- Visitor counts (daily, weekly, monthly)
- Dwell time, visit frequency, and peak hours
- Conversion rates when paired with POS data
- Cross-shopping patterns and customer origins
- True trade area boundaries
How it is collected:
- Mobile device GPS data, modeled up from a device panel
- In-store sensors (thermal, infrared)
- Wi-Fi and Bluetooth tracking
- Video analytics
- Point-of-sale integration
Most retail transactions still happen in stores, and the store footprint moves digital sales too: ICSC's halo effect research found that opening stores lifts online sales by 6.9% in a market, while closing them cuts online sales by 11.5%. A single bad location can cost $7-10 million over a lease term, which is why the question "where should I open next?" is worth answering with evidence.
I'm Clyde Christian Anderson, Founder and CEO of GrowthFactor.ai. After years in my family's retail business and in investment banking, I started GrowthFactor to solve the industry's biggest problem: picking the right locations with confidence instead of guesswork. Customers like Cavender's Western Wear used this approach to open 27 new stores in 2026, up from 9 in 2024, while Books-A-Million saves 25 hours per week per analyst.
Further reading: footfall analytics · foot traffic analysis · foot traffic data provider comparison
Retail foot traffic statistics for 2026
U.S. retail foot traffic rose 2.0% year over year in June 2026 while retail sales grew 8.4%, the fastest annual sales growth since 2022, per Colliers Retail Market Intelligence (published July 20, 2026). Visits are growing far slower than dollars, which means most of the sales gain is price and basket size rather than more people walking in.
Where those visits went in June 2026, per the same Colliers report:
| Category | Year-over-year visit change |
|---|---|
| Discount and dollar stores | +9.2% |
| Clothing stores | +4.9% |
| Hobbies, gifts, and crafts | +28% |
| Theaters and music venues | +27.4% |
Shopping center formats tell a similar story. Placer.ai's June 2026 Mall Index reported first-half 2026 visit growth of 4.7% for open-air shopping centers, 1.9% for indoor malls, and 1.0% for outlet malls, with June marking the third consecutive month of growth across all three formats. Placer.ai also found the median household income of the captured market slipped slightly year over year, a sign that the gains come from a broader mix of shoppers rather than a higher-income surge.
Two practical readings for a real estate team. First, open-air convenience-driven centers are the format with momentum, and they have set the pace every month of 2026. Second, a category that grows visits while the market grows 2.0% is taking share from somewhere, so category benchmarks matter more than market averages when you underwrite a site. For longer-run pattern context, see our guide to retail foot traffic trends.
Key metrics revealed by foot traffic data
Foot traffic data is only useful once you know which number answers which question. Visit counts tell you about reach, dwell time tells you about engagement, and origin data tells you where a trade area actually ends.
- Visitor count: total people entering a store, an indicator of reach and marketing effect.
- Dwell time: how long visitors stay. Placer.ai data puts Costco visitors at an average of 37.3 minutes against Walmart at 31.8 and Target at 28.7, a gap that tracks with basket size.
- Conversion rate: the share of visitors who buy, calculated by comparing traffic to POS data.
- Peak hours: the busiest times of day and week, used for staffing and inventory.
- Visit frequency: how often customers return, the clearest loyalty signal in the dataset.
- Cross-shopping: where customers go before and after your store, useful for cotenancy decisions.
- True trade area: where customers actually originate, rather than a three-mile ring drawn on a map.
More detail on each in our footfall analytics guide.
How to collect and analyze retail foot traffic data
No single collection method answers every question. Panel data is built for market-level questions like trade areas and competitors; sensors and video are built for in-store questions like layout and conversion. Pick the method that matches the decision.
| Method | Pros | Cons | Best use case |
|---|---|---|---|
| Mobile device data (GPS) | Broad view of trade areas and competitors; large privacy-compliant samples; tracks journeys beyond the store | Sampling bias; needs heavy processing; weak for in-store movement | Site selection, trade area analysis, competitive benchmarking |
| People counting sensors | Highly accurate at entry points; integrates with internal systems | Limited coverage; expensive to scale; no movement detail | Entrance counts, department conversion, zone staffing |
| Wi-Fi/Bluetooth tracking | In-store heatmaps; zone dwell time; new vs returning | Requires opt-in; privacy handling; imprecise counts | Store layout, aisle flow, in-store promotions |
| Video analytics | Rich movement and behavior data; doubles as loss prevention | High cost; significant privacy exposure; complex to run | Behavioral analysis, demographic appeal, theft prevention |
| POS integration | Links visits to sales; easy to integrate | Only counts buyers; no movement or dwell insight | Conversion rates, traffic-to-sales correlation |
Whichever mix you choose, the analysis discipline is the same: keep collection consistent across locations so comparisons hold, track a short list of KPIs tied to a business decision, read trends across seasons rather than single weeks, and normalize for holidays, weather, and local events before you call a change real. We compare the tooling in our location intelligence tools guide and the vendors themselves in our foot traffic data provider comparison.
How accurate is foot traffic data, really
Mobile foot traffic data is a modeled estimate, not a headcount. A provider observes a panel of devices that opted into location sharing, attributes pings to a store polygon, and extrapolates that sample to the full population using demographic weighting. Every step in that chain adds error, and the error is not random.
The published research is blunt about it. A 2023 PLOS One study of mobile location data across spatial scales found panels underrepresent Hispanic populations, low-income households, and people with lower education levels, while over-representing higher-income groups, with average sampling rates around 7.5% of the population. On the vendor side, Placer.ai reports correlations "consistently exceeding 90%" against first-party and authoritative sources, per a case study published by the Urban Libraries Council. SafeGraph publishes a 95%+ accuracy figure for its POI dataset, which is a claim about place records being correct, not about visit counts being correct. Those two numbers get conflated constantly, including by people selling you data.
Here is the position we take at GrowthFactor, and it is not the flattering one for a company that ships foot traffic in its product: third-party foot traffic is a relative signal, not ground truth. It is reliable for ranking a corner against the corner across the street, comparing cotenant draw within one trade area, and sizing competitive overlap. It is not reliable as the basis for an absolute sales forecast, and it degrades badly in vertical malls and dense strip centers where GPS drift and polygon misattribution put visits in the wrong tenant. Absolute forecasting belongs to your own store performance data.
That leads to four practices worth adopting before you underwrite anything on panel data:
- Check POI fidelity. Confirm the store polygon fits the real footprint, including multi-tenant parcels and stacked levels, and that visit rules exclude drive-bys.
- Demand the weighting method. Ask how the panel is post-stratified to local demographics and device penetration. If the vendor will not describe it, that is your answer.
- Calibrate against ground truth. Line up mobile-derived visits with door counters and POS for stores you already operate, then carry that correction factor into new markets.
- Set a minimum sample. Thin device counts at a location produce numbers that look precise and mean nothing.
This is also why our scoring is transparent by design. Every site score shows the inputs that moved it, from foot traffic to demographic fit to competition, across whatever lenses your workspace runs, so a number you cannot defend never leaves the room.
How to measure marketing performance with foot traffic data
Foot traffic is the only widely available way to connect advertising spend to physical store visits, which is why marketing teams now buy this data as often as real estate teams do. The measurement question is different from the site selection question: you are not asking how busy a place is, you are asking whether your campaign made it busier.
Correlation will mislead you here. Visits rise in December whether or not you ran ads. The methods that survive scrutiny are causal ones:
- Matched-market tests: run the campaign in one set of markets, hold out comparable markets, and read the difference in visit lift.
- Difference-in-differences: compare the change in exposed store visits to the change in unexposed stores over the same window.
- Geofenced exposure windows: measure visits from device populations exposed to an ad against a comparable unexposed population.
Report incremental visits, not total visits. A campaign that coincided with 40,000 visits and drove 1,200 of them is a 1,200-visit campaign. The same discipline applies to in-store merchandising: track traffic to a promotional zone against a baseline period to see whether a display moved anyone. Dwell time and conversion, read together with POS, tell you whether the traffic you bought was the traffic you wanted.
Privacy rules for location data changed in 2026
If you buy mobile location data, the legal ground under it shifted this year, and diligence on your vendor's consent chain is now part of the purchase. Several U.S. states moved from regulating location data to banning its sale outright.
Oregon's ban on selling precise geolocation data took effect January 1, 2026, and Virginia became the third state to ban the sale of precise geolocation data effective July 1, 2026, with Connecticut's amendments following on October 1, 2026. California's Delete Act now lets residents send a single deletion request that reaches every registered data broker, and brokers must process those requests starting August 1, 2026. The FTC has been active on the same front, settling its long-running case against location data broker Kochava in May 2026 on terms requiring affirmative express consent before sensitive location data is sold.
Practically, that means three questions for any provider: where does the consent come from, which states are excluded from the panel and how does that change your coverage, and how is the data aggregated before it reaches you. A provider who cannot answer the second question is telling you their coverage map is about to change without warning.
Using foot traffic data for site selection
For a new location, foot traffic answers questions a demographic report cannot: who is already moving through this parcel, where do they come from, and which of your existing stores will lose them.
- True trade areas: derive the catchment from observed customer origins and cluster home and work locations into primary and secondary rings, rather than drawing a radius.
- Cannibalization exposure: measure how much of a candidate site's draw already belongs to your own stores. Gravity models such as the Huff model estimate the share shift when a new store opens.
- Cotenancy and cross-shopping: find which neighbors actually deliver your customer, and which just look good on a site plan.
- Analog benchmarking: compare a candidate's traffic profile against your best and worst existing stores, then let your own sales data set the forecast.
Results from teams working this way:
| Customer | Result | How they did it |
|---|---|---|
| Cavender's Western Wear | 27 new stores in 2026, up from 9 in 2024 | Data-driven site selection with transparent scoring |
| Books-A-Million | 25 hours saved per week, per analyst | One consolidated platform replacing spreadsheets |
| TNT Fireworks | 10x more sites reviewed per cycle, 150+ locations opened in under six months | Automated screening ahead of human review |
More detail in our data-driven site selection and retail site selection analysis guides, and on forecasting in our predictive retail analytics guide.
Frequently Asked Questions about Retail Foot Traffic
What are the latest retail foot traffic statistics?
U.S. retail foot traffic rose 2.0% year over year in June 2026 while retail sales grew 8.4%, per Colliers Retail Market Intelligence (published July 20, 2026). Value and experience categories led: discount and dollar store visits were up 9.2%, clothing store visits up 4.9%, and theater and music venue traffic up 27.4%. Across shopping centers, Placer.ai's June 2026 Mall Index reported first-half 2026 visit growth of 4.7% for open-air centers, 1.9% for indoor malls, and 1.0% for outlet malls.
How accurate is retail foot traffic data?
Mobile foot traffic data is a modeled estimate, not a headcount. Providers observe a phone panel that covers a single-digit percentage of visitors and extrapolate to the full population using demographic weighting. A 2023 PLOS One study of mobile location data found panels underrepresent lower-income, Hispanic, and less-educated populations, so unweighted numbers carry real bias. Placer.ai reports correlations above 90% against first-party and authoritative sources when properly validated. Treat foot traffic as a strong relative signal for comparing sites, corners, and cotenants, and use your own door counts and POS data for absolute forecasts.
How do you measure retail foot traffic?
Retail foot traffic is measured through several methods: mobile GPS data captures device signals to count visitors and track origins, which is the most common method for site selection. Infrared and thermal sensors count people at entrances. Wi-Fi and Bluetooth tracking monitors device presence for dwell time and movement patterns. Video analytics provides detailed behavioral data. POS integration connects visits to purchases for conversion rate calculation. Most retailers combine methods, using panel data for market-level questions and sensors for in-store questions.
How does GrowthFactor compare to Placer.ai for retail foot traffic analysis?
Placer.ai offers the deepest foot traffic dataset in the market, with granular visitor counts, dwell time benchmarks, and cross-shopping behavior that are genuinely industry-leading for pure traffic analysis. GrowthFactor integrates foot traffic data (via Unacast) into a complete site selection workflow that includes scoring, demographics, competitive mapping, cannibalization modeling, and deal pipeline management in one platform. Placer.ai gives your team the raw traffic data to analyze; GrowthFactor gives your team the analysis itself with every variable visible. Lil Sweet Treat's two-person real estate team evaluates 120+ sites per month using GrowthFactor without any additional analyst headcount.
What is the difference between GrowthFactor and Esri for retail location intelligence?
Esri ArcGIS is the most powerful GIS platform available, offering maximum flexibility for teams with dedicated GIS analysts who can build custom spatial analyses. GrowthFactor is purpose-built for retail site selection teams that need answers without GIS expertise. Where Esri requires specialized knowledge to configure layers, run queries, and interpret spatial data, GrowthFactor delivers a scored site report in seconds from any address. Both platforms serve real estate teams, but the key difference is who does the work: with Esri, your analyst builds the analysis; with GrowthFactor, the platform builds it and your team interrogates the results. Cavender's Western Wear tripled their new store openings from 9 to 27 per year after switching to this self-serve scoring approach.
Where to start
Foot traffic data earns its keep when you are honest about what it is: a modeled estimate good at ranking options and sizing overlap, paired with your own sales data for anything absolute. Teams that treat it that way move faster because they stop arguing about whether the number is right and start arguing about what to do with it.
If you want the traffic, demographics, competition, and zoning for a site in one view, with every input behind the score visible, see our all-in-one real estate platform for retail.