Composite example (based on common failure modes): A regional coffee chain opened a new location that looked perfect on paper β great demographics, strong daytime traffic, growing suburb. It underperformed immediately. The post-mortem found a simple issue: a freight rail line and limited crossings turned "nearby" neighborhoods into a long detour. A 3-mile radius had counted those households as reachable. A 10-minute drive-time trade area would have excluded them on day one.
That's the core trade-area failure: counting people who can't realistically reach you when it matters. Any barrier that creates detours - rail lines, rivers, limited-access highways - breaks radius assumptions.
1. What Is a Trade Area?
The geographic region from which a retail location, restaurant, or service business draws the majority of its customers.
A trade area is not a perfect circle. It's not a ZIP code or a city boundary. It's the actual shape of where your customers come from β influenced by roads, traffic patterns, competitors, and the friction of distance.
Think of your trade area like a magnetic field around a magnet. The pull is strongest at the center and weakens as you move outward. But unlike a magnet, your field isn't uniformβit stretches along highways that speed access, and compresses where rivers, railroads, or congestion create barriers.
Understanding your trade area answers the most fundamental question in site selection: Who can you reach?
Throughout this guide, we'll follow a hypothetical coffee chain evaluating a new location in a competitive suburban market. Each section shows how the concepts apply to a real decision. Details are composites based on common scenarios.
For this suburban site, the trade area looks very different depending on how you define it. In some suburban markets, a radius-based trade area can overcount reachable population by tens of thousands versus peak-hour drive time. That's not a rounding error; it's the difference between a go and a no-go decision.
The size and shape of any trade area depend on:
- Your concept - Convenience goods (groceries, gas, coffee) have smaller trade areas than destination goods (furniture, specialty retail)
- Your location type - Urban, suburban, and rural trade areas look fundamentally different
- Competition - Nearby alternatives compress your reach
- Transportation infrastructure - Highways extend reach; rivers and dead-ends constrict it
A trade area isn't where customers could come from in theory β it's where they will come from given real-world constraints. The companies that get this right use travel time, not straight-line distance.
2. Why Trade Areas Matter
Trade area analysis is the foundation of every site selection decision. Get it wrong, and everything downstream β your demand estimates, competition counts, demographic analysis β will be wrong too.
Trade area analysis enables you to:
Estimate demand accurately
By aggregating demographics (population, income, households) within your true catchment β not an arbitrary radiusβyou get a realistic picture of how many potential customers exist.
When our hypothetical chain aggregated demographics within a 10-minute drive-time (vs. a 3-mile radius), median household income dropped from $112K to $94K. The closer-in population skewed younger and more apartment-heavy. Same site, different method, materially different customer profile.
Measure competition meaningfully
A competitor 3 miles away might be 25 minutes in traffic. Another 5 miles away might be 8 minutes on the highway. Distance alone doesn't capture the competitive landscape; travel time does.
Prevent cannibalization
For multi-unit operators, overlapping trade areas between your own stores dilute performance. Trade area analysis quantifies the overlap before you sign the lease.
Defend decisions in committee
When the CFO asks "why this site?"βyou need more than intuition. Trade area analysis provides the auditable methodology that survives scrutiny.
3. Primary, Secondary & Tertiary Trade Areas
Trade areas are typically segmented into three tiers based on customer concentration. Think of it like fishing in a pond: the primary area is where the fish are biting; the secondary is where you might get lucky; the tertiary is casting into the weeds.
Primary Trade Area
The geographic region producing 50-70% of sales or visits. These are your most frequent customers β the ones who chose you because you're convenient. The primary trade area is where marketing investments yield the highest return.
Typical benchmarks (vary by market and concept): QSR and coffee primary trade areas often extend 5-7 minutes drive-time; grocery, 10-12 minutes; specialty retail, 15-20 minutes. Validate against your own performance data where available.
Secondary Trade Area
The surrounding region contributing 15-25% of business. Customers here visit less frequently, often because you offer something competitors don't. Combined with the primary area, this forms your "Main Trade Area" (MTA).
Tertiary Trade Area
The fringe, capturing 5-15% of remaining customers. These are destination shoppers, tourists, or business travelers. For most concepts, the tertiary area is too diffuse to target efficiently β but for destination retailers, urgent care clinics, or unique dining concepts, it can be substantial.
Customer value decays with distance. The further customers travel, the less frequently they visit. A household 5 minutes away might visit weekly; a household 20 minutes away, monthly. This is why primary trade area definition is so critical β it's where the repeat business lives.
4. Methods: Radius vs. Drive-Time vs. Observed
There are three fundamental approaches to defining trade areas. Each has its place β but they are not equally accurate.
Draw a circle of X miles around a location
The simplest approach: 1-mile, 3-mile, and 5-mile rings around a site. Easy to understand, easy to generate, and useful for quick comparisons between markets.
Strengths
- Fast and easy to generate
- Good for initial screening
- Apples-to-apples market comparisons
Limitations
- Ignores road networks
- Ignores traffic and travel time
- Ignores barriers (rivers, highways)
- Treats all directions equally
Define boundaries by travel time: 5, 10, 15 minutes
Instead of distance, use time. A 10-minute drive isn't a circle β it's shaped by road networks, speed limits, and traffic patterns. Isochrones (from Greek: iso = equal, chronos = time) capture this reality.
Strengths
- Accounts for actual road networks
- Can incorporate traffic data
- Much more realistic than radius
- Works for drive, walk, transit modes
Limitations
- Still a model, not observed behavior
- Requires traffic-aware routing engine
- Time of day affects accuracy
Define boundaries based on where customers actually come from
The "ground truth" approach. Using actual customer data (addresses, mobile device pings, loyalty data), you can see exactly where your visitors originate.
Strengths
- Based on real behavior, not models
- Accounts for all factors automatically
- Most accurate for existing locations
Limitations
- Requires existing location or proxy
- Not available for greenfield sites
- Mobile data can have accuracy issues
For a greenfield site, use drive-time isochrones for initial analysis. Then validate by pulling mobile foot traffic data from a nearby competitor as an analog β seeing where their visitors actually come from. You may find the trade area extends further in directions with less competition, or compresses where alternatives cluster.
| Method | Best For | Accuracy | Data Required |
|---|---|---|---|
| Radius | Quick screening, market comparisons | Low | Address only |
| Drive-Time | Site evaluation, greenfield analysis | Medium-High | Address + routing engine |
| Observed | Existing locations, analog modeling | Highest | Customer data or foot traffic |
Use radius for quick back-of-the-napkin screening β but switch to isochrones before committing real money. And whenever possible, validate your models against observed foot traffic data.
5. Deep Dive: Isochrones
An isochrone is a polygon representing all points reachable within a given travel time from a starting location. Unlike a radius circle, an isochrone follows the actual road network. It bulges along highways and contracts where roads are slow or sparse.
Rule of thumb: Use a radius for a rough picture, but switch to isochrones before you commit real money. The extra analytical rigor pays for itself when it catches a site with access problems that simple geometry misses.
How Isochrones Are Generated
- Start with a road network graph - Streets become edges with speed attributes; intersections become nodes
- Run shortest-path algorithms - Calculate travel time from the origin to all reachable points
- Draw the boundary - Connect all points reachable within the time threshold into a polygon
Traffic-Aware Isochrones
Static isochrones assume free-flow traffic. But a site's 10-minute trade area at 2pm looks different than at 8am or 6pm. Traffic-aware isochrones incorporate historical or real-time traffic patterns to show how catchments shift throughout the day.
Many tools generate isochrones using straight-line distance decay, not actual routing. Always verify your isochrone provider uses real road network data and (ideally) traffic patterns. An isochrone that crosses a river with no bridge is a sign of poor data quality.
Standard Time Thresholds
- Walk-time: 5, 10, 15 minutes (at 3-3.5 mph average walking speed)
- Drive-time: 5, 10, 15, 20 minutes
- Transit-time: 15, 30, 45 minutes (highly schedule-dependent)
Three bands is the sweet spot for most analyses. More than four gets visually cluttered; fewer loses nuance.
See isochrones in action
Drop any address into Geod and get time-aware trade areas, demographics, and competition density in seconds.
Try Free - No Card Requiredβ6. Time-of-Day Variation
One of the most overlooked factors in trade area analysis: your trade area changes shape throughout the day.
Rush hour traffic compresses catchments. Weekend patterns differ from weekday patterns. A breakfast spot's 7am trade area is fundamentally different from a bar's 10pm trade area.
Why This Matters
- A site that looks great at 2pm on Tuesday might be inaccessible when your customers actually show up
- Weekend vs. weekday traffic patterns can materially shift reachable population
- Daytime population (workers) vs. nighttime population (residents) serve different concepts
Composite example: Running isochrones at 7am (rush hour) vs. 11am (off-peak) for the same site showed the 10-minute polygon shrinking by 34% during rush hour due to highway congestion. For a morning-focused concept like coffee drive-thru, the rush-hour analysis is the relevant one β revealing fewer reachable households than off-peak numbers suggest.
Best Practice
Run isochrone analyses at multiple times of day that match your peak business hours. A QSR focused on lunch should care about 11am-1pm accessibility. A sit-down dinner restaurant should care about 6-8pm.
Static trade areas lie. A site's catchment at 2pm on Tuesday tells you nothing about accessibility during your actual peak hours. Always model for the times that matter to your concept.
7. A Brief History of Trade Area Analysis
The science of trade area analysis has evolved significantly over the past century. Understanding this history helps contextualize why modern methods exist.
8. Gravity Models Explained
Gravity models are the theoretical backbone of trade area analysis. They formalize the intuition that larger, more attractive stores draw customers from further away β but that distance creates friction.
Reilly's Law (1931)
Two cities attract trade from an intermediate town in proportion to their populations and inverse proportion to the squares of their distances.
Useful for: Understanding inter-city retail competition; calculating the "breaking point" between two retail centers.
Huff Model (1964)
The probability that a consumer at location i will visit store j depends on the store's attractiveness divided by travel time (raised to an exponent), relative to all alternatives.
The Huff model produces probabilistic trade areas β surfaces showing the likelihood of a customer choosing your store over competitors. Unlike hard boundaries, these capture the reality that trade areas overlap.
Source: Huff, D.L. (1964). "Defining and Estimating a Trading Area." Journal of Marketing, 28(3), 34-38.
Modern Applications
While pure gravity models have limitations (they assume rational consumers with perfect information), they remain foundational. Modern site selection tools often combine Huff-style probability models with observed foot traffic data to calibrate and validate predictions.
9. Data Sources for Trade Area Analysis
Accurate trade area analysis requires layering multiple data sources. Here are the primary categories:
Demographics
Who lives within the trade area? The standard source in the United States is the American Community Survey (ACS), published by the U.S. Census Bureau. ACS provides:
- Population and households
- Median and average household income
- Age distribution
- Educational attainment
- Commute patterns
ACS data is released at multiple geographic levels: state, county, tract, and block group. For trade area analysis, block group (the most granular public level) is typically most useful.
Data vintage matters. ACS 5-year estimates (e.g., 2019-2023) provide stability; 1-year estimates provide recency. Always document the vintage in your analysis.
Competition / Points of Interest (POI)
Who else is serving this trade area? POI databases catalog business locations with category tags, brand names, and addresses. Major providers include:
- Foursquare Places (100M+ global POIs, updated monthly)
- Precisely
- Google Places
Quality varies significantly. Look for: recency of updates, category accuracy, and coverage completeness.
Foot Traffic / Mobility Data
Where do customers actually come from? Mobile location data, derived from smartphone GPS pings (with user consent), provides observed trade areas. Providers include:
- Placer.ai
- Unacast (vendor-reported validation: ~0.93 correlation with ground truth)
- Advan (via resellers such as Dewey Data)
Foot traffic data provides "ground truth" validation for modeled trade areas, and is especially valuable for analyzing competitor locations. Note: the mobility data market has consolidated significantly since 2022 β verify vendor viability before committing.
Caveats to keep in mind:
- Sampling bias β devices β people; panel composition varies by demographics
- Dwell-time filtering β pass-through traffic may be included or excluded depending on methodology
- Privacy/aggregation thresholds β low-traffic locations may lack sufficient data
- Best use β validation and analog analysis, not as an oracle
Geographic Boundaries
For spatial analysis, you need authoritative boundary files:
- TIGER/Line (Census Bureau) - Census tracts, block groups, ZIPs, counties
- OpenStreetMap - Road networks for routing
- State/Local sources - Parcels, zoning, jurisdictional boundaries
License & Permit Records
State-level license records (liquor licenses, health permits, professional licenses) can validate business operating status and provide competitive intelligence not available in commercial POI databases.
10. Cannibalization Analysis
For multi-unit operators, the question isn't just "is this a good site?" β it's "is this site net additive to our network?"
When a new location draws sales from existing stores in the same network, rather than capturing net new demand. The phenomenon of "stealing from yourself."
Why It Matters
Opening a store 2 miles from an existing high-performer might look attractive in isolation. But if the trade areas overlap significantly, you're not expanding your reachβyou're splitting existing customers between two locations. The result: lower performance at both stores.
Composite example: A chain with an existing downtown location considers a new site 4 miles away. Trade area overlap analysis reveals that ~30% of the new site's 10-minute catchment overlaps with the existing store. Net new reach is materially lower than gross numbers suggested β still potentially viable, but the adjustment changes the pro forma.
How to Measure Cannibalization
1. Trade Area Overlap
Calculate the percentage of the new site's trade area that intersects with existing stores. A 30% overlap is very different from a 5% overlap.
2. Demand Transfer Estimate
Within the overlap zone, estimate what share of demand will transfer from existing stores to the new one. This depends on relative accessibility and site characteristics.
3. Net New Demand
Subtract estimated transfer from gross opportunity:
Example: A new site shows 45,000 population in its trade area. But 8,100 of those residents (18%) already shop at your Store #47, and would likely switch. Net new demand: 36,900 (82%).
Best Practice
Err on the side of conservative (pessimistic) transfer estimates. Better to reject a borderline site than to open and cannibalize.
Single-site analysis is necessary but not sufficient for network operators. Every new site must be evaluated against the existing portfolio. A "great" site that cannibalizes 40% of a nearby store's revenue isn't great β it's a wash.
11. Industry-Specific Applications
Trade area dynamics vary significantly by industry. Here's how the principles apply differently:
Retail Trade Area Considerations
Convenience vs. Destination: A grocery store (convenience) draws 70%+ of customers from within 10 minutes. A furniture store (destination) might draw 60% from 20+ minutes. Define your category before setting thresholds.
Co-tenancy Effects: Anchor tenants (Target, Whole Foods) expand your effective trade area by drawing traffic you wouldn't generate alone. Analyze co-tenant foot traffic, not just your category.
Example Thresholds:
- Grocery/Drug: 5-10 minute primary
- Apparel/Soft goods: 10-15 minute primary
- Furniture/Home: 15-25 minute primary
Restaurant Trade Area Considerations
Daypart Complexity: Restaurants often serve multiple trade areas: breakfast draws from commuters (work-based), lunch from nearby workers (hyper-local), dinner from residents (residential).
Daytime vs. Nighttime Population: A downtown lunch spot should analyze daytime population (office workers), not residential census data. These can differ by 5-10x.
Pass-By Traffic: QSR and fast-casual depend heavily on vehicular pass-by traffic. A location with 30,000 ADT (average daily traffic) on the adjacent road may outperform one with 50,000 population in the radius but poor visibility.
Example Thresholds:
- QSR (drive-thru): 5-7 minute primary
- Fast-casual: 7-12 minute primary
- Casual dining: 12-20 minute primary
- Fine dining: 20-30+ minute (destination)
Healthcare Trade Area Considerations
Acuity-Based Trade Areas: Patients will travel much further for specialized care than for primary care. An urgent care clinic has a 10-15 minute trade area; a regional cancer center may draw from 60+ minutes.
Referral Patterns: Unlike retail, healthcare often involves referral networks. A specialist's trade area is partly defined by which PCPs refer to themβa non-geographic factor.
Access & Equity: Healthcare planners often analyze trade areas to identify underserved populationsβareas where residents lack access to care within acceptable travel times. This "void analysis" is the inverse of site selection.
Example Thresholds:
- Urgent care: 10-15 minute primary
- Primary care (PCP): 15-20 minute primary
- Specialty ambulatory: 20-30 minute primary
- Regional hospital: 30-45+ minute secondary
Benchmark: 85% of patients choose a PCP within 15 miles (Medical Group Management Association, 2022)
12. Best Practices for Trade Area Analysis
Match the method to your concept
- Convenience concepts (coffee, QSR, grocery) β Tight isochrones (5-10 min drive, 10-15 min walk)
- Destination concepts (furniture, specialty retail) β Wider isochrones (15-30 min drive)
- Urban locations β Walk-time or transit-time analysis
- Suburban/rural β Drive-time analysis
Use time-of-day appropriate windows
If your business peaks at lunch, analyze 11am-1pm accessibility. If it's dinner-focused, analyze 5-7pm. Static analyses miss critical patterns.
Document your methodology
Trade area definitions should be consistent across your organization. Document: time thresholds, traffic assumptions, data sources, and snapshot dates. This enables apples-to-apples comparisons and audit trails.
Validate models with observed data
If you have existing locations, compare modeled trade areas against actual customer origination. Calibrate your assumptions based on reality.
Aggregate demographics properly
Census data is published at fixed geographies (block groups, tracts). Trade areas don't respect those boundaries. Use area-weighted interpolation or population-weighted aggregation to estimate demographics within custom polygons.
Don't forget daytime population
Census data captures residential population. But for locations serving office workers (lunch spots, business services), daytime population may matter more than nighttime.
Revisit annually
Trade areas aren't static. New competitors, road construction, demographic shifts, and changing consumer behavior all affect catchments. Best-in-class operators refresh their trade area models at least annually.
Summary
Trade area analysis is the foundation of data-driven site selection. The key principles:
- Use isochrones, not radii - Travel time beats straight-line distance
- Account for time of day - Rush hour reshapes catchments
- Layer multiple data sources - Demographics + competition + mobility
- Quantify cannibalization - For networks, overlap matters
- Tailor to your industry - Coffee β furniture β healthcare
- Document methodology - Defensible decisions require audit trails
- Apply consistently - Same model, every site, every time
The value of this methodology isn't just better decisions - it's defensible decisions. When you can show time-aware isochrones, quantified cannibalization, and sourced demographics, the real estate committee has something to evaluate beyond intuition. Sites that pass this filter tend to perform closer to projections. Sites that would have failed it - but got approved anyway on gut feel - are where the expensive lessons come from.
The companies that systematize their location strategy β rather than relying on individual intuition β make faster, more consistent, more defensible decisions. They open more stores, with higher hit rates, and catch problems before committee instead of after opening.
That's what Geod is built for.
References
- Access Development. (2016). The Impact of Retail Proximity on Consumer Purchases. PDF
- Buxton. "Retail Trade Area Analysis: Data in High Demand by Developers." buxtonco.com
- Foursquare. "Places Overview." Foursquare Docs
- Huff, D.L. (1964). "Defining and Estimating a Trading Area." Journal of Marketing, 28(3), 34-38. SAGE Journals
- Reilly, W.J. (1931). The Law of Retail Gravitation. New York: Knickerbocker Press.
- Unacast. "An In-Depth Guide to Foot Traffic Data." unacast.com
- U.S. Census Bureau. "Quarterly Retail E-Commerce Sales Report." census.gov
- U.S. Census Bureau. "American Community Survey (ACS)." census.gov