Ideally, a trade area represents the spatial extent of a store’s customers and where they reside or originate. With this information, businesses can determine if an area can support more retail space and, based on demographics, lifestyles, and consumption patterns, whether a certain location type is needed to compete effectively.
The trade area’s appearance depends on the algorithm used to create its boundaries and the methodology used to represent the market. Trade area type differences affect usefulness for real estate and marketing purposes. Some are ideal for visual inspection and display, while others are more appropriate for data retrieval. The most common trade area analyses methods include rings, customer penetration polygons, drive-time/drive distance polygons, Voronoi diagrams and gravity methods.
Rings do a great job of quickly summarizing customer and market data around a site (see “In Defense of Rings,” page 20), but they don’t accurately represent the true trade area. Retailers and real estate companies frequently screen sites using ring-based “site reports” from demographics companies. These reports show primary, secondary and tertiary concentric rings with summarized market data within each ring.
The primary ring captures the nearest source of customers–local residents and employee base–as well as the nearest competitors and other demand generators. The secondary ring encompasses residential neighborhoods where most customers live. A tertiary ring identifies farther flung customers for destination businesses and major traffic corridors feeding the business.
Customer Penetration Polygons
Trade area analysis commonly is used to determine a single site’s influence area based on its draw from local workplaces, neighborhoods and commuter flows. More accurately reflecting a trade area than rings allow, polygons are drawn by accumulating customer households and plotting the outside boundary for a predetermined percentage of customers. It’s nearly impossible to plot all customers within a reasonably sized trade area, so the polygon usually traces the area from which 80%-90% of the customers or sales will originate. The other 10%-20% of the business comes from outside the trade area.
Penetration polygons are drawn as overlapping arcs for multiple sites. The amount of overlap is important for assessing cannibalization potential from existing sister stores. Using a neighboring competitor’s customer sample, marketers can assess the opportunity to steal those customers by plotting competitor location and proposed sites within the trade areas, and measuring the degree of infringement.
Penetration polygons illustrate store coverage relative to market and demographic characteristics, demographic variations, street networks, physical features and access barriers that stretch or shrink a trade area.
Drive-time and drive-distance polygons enhance trade area delineation by reflecting how far customers are willing to travel.
Drive-time trade areas use node and link versions of digitized street databases, rules for assigning speed limits and algorithms for determining shortest travel paths. Average drive speeds are assigned to different road classes. Marketers calculate point-to-point times using the nearest road junctions. Some approaches also interpolate exact travel times and use points in polygon algorithms to pull demographic data for specified areas.
The major advantage–drive-time polygons account for “pull” along major road networks. The most accurate approach requires customization of posted speed limits, particularly for local analysis. The primary limitation of drive-time analysis is temporal–accounting for variations in congestion and flow by day and time of day. In addition, an area’s dominant traffic generator may supersede smaller trade areas. In other words, drive times for mall-based tenants may be supplanted by the mall’s own trade area, because customers likely will shop multiple mall stores.
Suppose you’re a real estate analyst for a major retail organization seeking to backfill (add more stores until the market is saturated) or rationalize (evaluate a market to optimize chain performance by adding new stores or changing existing ones) a series of large markets. To start, you’re interested in the market’s store and competitor distribution, as well as each site’s area of influence. Voronoi diagrams use geometry to compartmentalize the entire market into connecting polygon trade areas, allowing planners to estimate each site’s customer volume. Sites covering large geographic regions may be under-serving their markets, while sites with small Voronoi assignments may be saturated. In addition, sites with common boundaries probably are competing for customers in the same area, signaling a need for closer examination of neighboring competitors.
Voronoi diagrams provide rough-cut representations of a market’s trade area, but are based on several assumptions. Pricing and merchandising variations also influence consumers to spend their money in several stores. Also, customers may visit different sites on different occasions. This behavior, however, can be interpreted with gravity models.
Analysis methods interpreting customer shopping behavior variations provide the most accurate trade area reflection. Generally, consumers don’t like to travel long distances, particularly for grocery shopping, banking or movie viewing. But polygon and ring-based trade areas don’t account for consumers choosing to shop elsewhere some or all of the time.
Another difficulty is in representing the area “probabilistically,” or as a surface in which competitors’ trade areas blend into each other and households fall into more than one area. Gravity and spatial interaction models further interpret customer behavior. Supply and demand mathematics help define household tendencies to patronize a particular store some of the time based on distance to the store and its attractiveness relative to other shopping options.
Gravity techniques examine each market’s demand and allocate that demand to existing service locations based on distance and attractiveness. Demand potential is determined by the area’s demographics and lifestyles, which can be tied to historical purchase rates and patterns for almost any product or service using syndicated commercial databases. The resulting trade area is a series of concentric rings or polygons in which each larger ring represents less business for the store, especially if attractive competing options are nearby.
Balancing the negative impact of increasing distance, store attractiveness can enhance draw. Attractiveness may be measured as square footage, a proxy for variety of goods sold, or as site/venue features, brand equity in the market, etc. Brand equity reflects “name recognition” in the market and can increase a store’s pull. Using sophisticated survey methods, brand equity is measured as the density of stores in the market or number of years the chain’s been in the market.
One large retailer uses an ongoing combined analysis in which customers relate what factors (store access, floor space, location, service, pricing, goods, promotions, etc.) are important. Optimal store profiles are generated and used to evaluate each store in the chain and to adjust its “attractiveness” levels for sales forecasting models.
Spatial interaction models also define sensitive trade areas that are susceptible to competitor activities. If strategically positioned competitors steal business, the trade area polygon shrinks.
Recently, San Diego-based National Decision Systems (NDS) helped a furniture retailer define gravity-based trade areas, using a comprehensive survey to measure store attractiveness and NDS’s MicroVision segmentation software to estimate a demand potential for each neighborhood. MicroVision can be linked to consumer expenditure potential estimates for any U.S. geography level, starting with ZIP+4.
A Trade Area Primer
Capturing Customer Data
In Defense of Rings