A trade area is the geographic region from which a business draws most of its customers. It isn't a circle selected in advance; customer data defines its boundary after you determine which customers to include and what distance is too far. Change either decision, and the boundary changes as well.
Two teams can review the same corner and reach different figures solely because they made those two choices differently. Neither team is lying.
What is a trade area?
Most retail chains still draw that boundary using a radius ring. But a 3 mile ring reveals very little about where your customers come from. It overlooks highways, rivers, competitor locations, and how people actually travel, presenting a guess as analysis.
And the consequences are material. Choose a site using a flawed trade area, and you commit to a lease and inventory for a location that was never going to meet its numbers. Repeat that across dozens of sites each year, and the error accumulates quickly.
I'm Clyde Christian Anderson, Founder and CEO of GrowthFactor.ai. Working in both retail and investment banking showed me firsthand how frequently a poorly defined trade area can derail an expansion plan. We built GrowthFactor for that reason, turning the work from guesswork into something a real estate committee can inspect.
"Trading area" and "trade area" are the same thing
A retail management course probably introduced you to trading area, while a data vendor called it trade area. The definitions are effectively the same. Older academic literature generally uses trading area; modern software and vendor materials generally use trade area. Both refer to the same boundary. Course material uses the two interchangeably inside a single chapter uses the terms without stopping to distinguish them. Use the wording your team already uses, and don't assume a broker deck is measuring something different because it uses another term.
The important distinction is between a trade area and a market area. A market area covers a product's full economic context within a geography, while a trade area is the portion tied to your particular store.
The three zones, and why the percentages decide the boundary
The traditional view separates a trade area into three tiers according to how much of your customer base each tier represents:
- Primary trade area. Your most frequent and geographically closest shoppers account for roughly 50 to 80% of your customers. The precise cutoff varies by source; retail management textbooks often define it as 55 to 70%.
- Secondary trade area. An additional 15 to 30% of the customer base. People in this group travel farther or shop less often, usually making a specific trip rather than taking a routine one.
- Tertiary or fringe trade area. The remainder consists of occasional customers, passersby, and the long tail of a destination store's draw.
Most teams interpret those percentages in reverse. They don't refer to a boundary that's already been drawn; they specify how to draw it. Rank customers by distance traveled, then cut the ranking at 60%. The shape enclosing the customers in that cut is your primary zone. Change the cutoff to 70%, and the zone expands, its demographics change, and any forecast based on those demographics changes with it.
The analog approach remains the most honest way to map a trade area when you have your own customer data.
Convenience versus destination changes everything
Convenience trade areas serve businesses that sell everyday necessities, including groceries, gas, and coffee. Since routine purchases don't send people driving far, these areas remain small and dense. Destination trade areas cover furniture, specialty apparel, entertainment, and anything else worth planning a trip around; at times, they extend across a whole metro.
Two locations within one brand can fall on opposite sides of this line. Because its assortment is deeper, a full-format flagship draws from farther away than a small-format express store designed for a five-minute stop. When both formats are in the portfolio, one trade area standard applied across the fleet overstates one and understates the other.
Trade area analysis in retail management: four models, four questions
Retail management courses teach four methods, but most vendor guides leave all four out. That is unfortunate: each method answers a different question, and selecting the wrong one produces a study that is precise yet beside the point.
Customer spotting and the analog method
It remains the oldest technique and the most direct one. Map customer origins with loyalty records, checkout ZIP codes, delivery addresses, or licensed device-origin data. Sort them by travel time and apply your thresholds. For the forecast, identify an existing store that matches the new site in format, road pattern, and competitive density, then use that store's sales per household.
It relies on your own customer data. In a market where you don't yet operate a store, it can tell you little beyond comparisons with analogous markets.
Reilly's law of retail gravitation
Published by William Reilly in 1931, this model draws directly on Newton. For the town between two retail centers, trade is pulled toward each center in proportion to its size and inversely to the square of its distance. Reilly expressed the relationship as a ratio of distances. The form people actually use is Converse's 1949 rearrangement, which calculates the breaking point in miles:
breakpoint distance from A = D ÷ (1 + √(size of B ÷ size of A))
where D measures the separation between the two centers, and size refers to population or square footage.
Its speed and need for almost no data explain why it has lasted. But Reilly's law makes severe assumptions: it models the map as flat and featureless, ignoring rivers, roads, and mountains that bend anyone's route, then divides the world with a firm boundary. Everyone on one side shops at A, while everyone on the other shops at B. Real shoppers don't act that way.
The Huff gravity model
David Huff's 1963 model resolved the hard-line problem by returning probabilities rather than a boundary. For a customer at any location, it estimates the likelihood of choosing your store over every alternative:
P(i chooses j) = (Aj^α ÷ Dij^β) ÷ the sum of (Ak^α ÷ Dik^β) across every competing store k
Attractiveness A usually refers to store square footage. D captures distance or travel time. The exponents α and β set the strength of size's pull and distance's push.
The model produces a probability surface over the full map, not a single line, and can account for many competitors at once. That is why the Huff model forms the basis of gravity modeling in most modern GIS site selection tools. Two cautions matter: the exponents must be calibrated using actual sales or foot traffic data, since an uncalibrated Huff model is a confident-looking guess. Size and distance alone provide only a thin explanation of why anyone picks a store. Modern practice adds demographics, competitive context, and observed foot traffic.
The index of retail saturation
The other three cover individual sites, while this one covers a market. Its index is the spending available in an area divided by the competing retail square footage pursuing it:
IRS = (households × annual category spend per household) ÷ competing retail square feet
A high number indicates unmet demand and available room. A low number means you'd be competing for a share in a market that's already full. It's the fastest initial screen before committing serious analysis time to a site, and it pairs naturally with market saturation analysis.
None of these four is self-executing. Each requires judgment calls, and you should record them before running it: which customers count, what distance is too far, and which competitors are actual competitors. If you skip that, you end up with a number nobody can defend in a committee room.
What actually changes a trade area's size and shape
Population density
In densely populated urban markets, a small radius contains many people, keeping trade areas compact. Customers have plenty of alternatives nearby and won't travel far for one retailer. In suburban and rural markets, the radius must extend much farther to reach the same number of customers because alternatives are fewer and longer drives are normal.
Road networks and transit
Highways extend a trade area in the direction they run. Even a short map distance can produce a smaller area when access is poor or congestion is persistent. Reliable transit expands it by bringing in customers who don't drive. That is the single biggest reason a ring and a drive-time polygon disagree, with the widest gap in the markets where the choice is hardest.
Physical barriers
Rivers, rail lines, highways without crossings, large parks, and industrial zones interrupt travel routes. When a river lacks a convenient bridge, it can eliminate half of a nominal trade area, although the map's ring continues to indicate that area is fully served.
Competitors and co-tenants
A strong competitor located between you and a neighborhood reduces your draw. Complementary businesses grouped together have the opposite effect, building a shared pull that expands every trade area in the district. For the same reason, lifestyle centers and outdoor malls create larger, more destination-style zones than neighborhood strip centers.
Traffic generators
Stadiums, universities, hospitals, and large office complexes draw people into an area, where they become customers of whatever is nearby. Daytime and residential populations can differ so significantly that an analysis based only on residents misses the actual customer base.
What the analysis is for
Site selection and expansion
This is the central step. Defensible trade-area boundaries help you identify markets where demand exceeds supply, assess a specific site based on the customers it can realistically reach, and determine whether a new store will cannibalize an existing one rather than expand the network. Our retail site selection process centers on this step, and site demographics only mean something once the boundaries they describe can be defended.
Forecasting
The trade area you draw determines the population and spending power credited to a site, so it underpins any top-down sales forecast. Set the boundary too broadly and you overstate the customer pool. Set it too narrowly and the model excludes revenue that exists. Either mistake produces a forecast the CFO can't reconcile with results a year later.
Marketing spend
Once you understand where customers originate, you can direct the budget there rather than spread it across the metro. The boundary used to qualify a site also identifies which ZIP codes warrant a mailer and which you've been paying to reach unnecessarily.
Franchise territories
Trade areas give franchisors the foundation for fair territory grants. Because each franchisee needs a viable customer base without competing against the brand next door, franchise territory mapping starts with trade areas rather than a map of counties.
Closure and relocation
Trade areas deteriorate as population moves, competitors open, and demographics shift. When a catchment no longer supports a store's cost structure, trade area analysis identifies the problem. It will sometimes also show that one of your nearby stores is already serving the same customers more effectively.
What changed since this was standard practice
Three shifts matter for anyone doing this work today.
The trip is no longer the only transaction. Buy online, pick up in store now exceeds delivery in grocery: 31% of shoppers use pickup against 29% using delivery, according to FMI and NielsenIQ data cited by eMarketer. The trade area now includes people who bought from home and drove to you solely to collect their order. Parking, curb access, and pickup convenience are therefore part of its boundary in ways they never were before.
Where you can buy location data is narrowing. Several states now prohibit outright the sale of precise geolocation data. In Virginia, the ban took effect on July 1, 2026, defining precise geolocation as anything that identifies someone within a 1,750 foot radius and applying to sales made for money. Connecticut's version takes effect on October 1, 2026. As a result, raw ping data is becoming harder to purchase, while vendors' licensed, aggregated products continue working fine. If your trade area method relies on device origins, ask your data provider what they're licensed to sell you and where.
The tooling has reshuffled. SiteZeus relaunched its platform under Atlas in May 2026, and Buxton now uses the Audiense name after consolidating its brands. Notes from a category evaluation conducted over a year ago already contain names that have changed.
Frequently Asked Questions about trade area analysis
Is a trading area the same thing as a trade area?
Yes. They are the same concept under two spellings. Retail management textbooks and older academic work tend to say trading area, while data vendors and modern site selection software say trade area. Nobody in the field treats them as different things, so if a broker deck says trading area and your software says trade area, they are describing the same boundary and you can compare the two directly.
What is the difference between a trade area and a market area?
A trade area is customer-centric: the geographic region your specific location draws its patrons from. A market area is the broader economic picture, covering every potential customer, competitor, and demographic trend for a product or service in that geography. One market area contains many trade areas, which is why a market can look healthy on paper while a particular site inside it has a weak trade area.
Which trade area model should I use for a new site?
Start with what you are asking. If you want to know where your customers actually come from, use customer spotting on your own store data. If you want to know how a new site will split demand with competitors, use a Huff gravity model. If you want a fast boundary between two towns with almost no data, use Reilly's law. If you want to know whether a market already has enough stores, use the index of retail saturation. Picking the model after you write down the question is the part most teams skip.
What is the biggest mistake in trade area analysis?
Drawing an arbitrary ring and calling it analysis. A 1, 3, or 5 mile radius assumes every direction is equally reachable, which ignores highways, rivers, one-way road patterns, and the competitor sitting between you and half your circle. The second mistake follows from the first: changing the method partway through a search. A ring on one site and a drive-time polygon on the next produces two numbers that cannot be compared, and that is a different failure from either method being wrong.
How does GrowthFactor compare to Placer.ai for trade area analysis?
Placer.ai's device panel is the category benchmark for foot traffic depth, providing granular visitor origin data that helps define where customers actually come from. GrowthFactor combines licensed foot traffic data with demographics, vehicle counts, and competition analysis, then layers trade area insights into a site scoring and deal management workflow. TNT Fireworks screened locations 60% faster using GrowthFactor's integrated trade area tools.
Where to take this next
The models above reflect a century of retail thought on one issue: how far someone will travel and for what reason. Better data hasn't made any of them obsolete. It has changed how quickly you can calibrate them and how candidly you can check them against what customers actually did.
The mistake to prevent isn't choosing the wrong model; it's applying four different definitions to four sites and comparing the outputs as though they were equivalent. Before the search begins, document the method, threshold, and competitor set. That gives you a trade area you can defend when the committee asks how the number was derived.
GrowthFactor puts that step in site selection instead of relegating it to a consultant's appendix, so the people who must sign the lease can see the boundary behind a score.