11.8: Site Location Analysis

Site Location Analysis

As the saying goes, “the three most important words in business are ‘location, location, location!’” Finding an ideal site for a new business is a key element in the success, or failure, of many businesses. Site location analysis is the primary job of geographers working in location intelligence jobs; a major source of high-quality employment for those with a geography degree. Many companies, including McDonald’s, Kohl’s, Wal-Mart, and Walgreens employ geographers to help them select optimal site locations for both retail and warehousing operations. Governments also need geographers to select ideal sites for schools, police departments, airports and fire stations. Sometimes geographers help decide where to close a business or public service facility. The process can be quite complex because multiple factors must be considered at each location. GIS is an indispensable tool for analyzing the interplay of multiple factors simultaneously. A site location analyst analyzes things like traffic patterns, real estate costs, zoning laws, economic competition, as well as the socio-economic, ethnicity and age structures of nearby customers/citizens among other things. Doing fieldwork to collect data about potential site locations is another important step in the decision-making process of location analysts.

Hotelling Model

Harold Hotelling developed the most basic site location model in the late 1920s. It is useful for understanding some basic patterns of retail in many cities. Hotelling’s basic premise was that when competing firms sell a similar product, customers will travel the shortest distance possible to purchase that product. Since competitors frequently sell products that are virtually indistinguishable from each other (gasoline, aspirins, Coca-Colas, etc.), the behavior of customers creates an incentive to agglomerate at a point that maximizes the potential number of customers.

Figure Infographic. The logic of the Hotelling Model plays out in a series of moves and counter-moves by competitors arranged in a linear market, like a street or on a beach. Eventually agglomeration will occur.

Suppose there are two competing firms on Main Street in some small town. Each firm has an incentive to capture the market share of its competitor by moving into the territory of the other. To capture the maximum number of customers, Firm A (Amy in the graphic) should move next door to Firm B (Joe), thereby making Amy’s store an intervening opportunity. By moving, Firm A (Amy) will capture the maximum number of customers traveling along Main Street. If Firm B (Joe) is smart, he will leapfrog Firm A (Amy), thereby capturing the maximum number of customers. Under this logic, after a series of moves and counter moves, a state of equilibrium will be reached that finds Firm A and Firm B located adjacent to each other, with each firm capturing nearly fifty percent of the customers. Of course, most businesses try to distinguish themselves from their competition through price, service, and product. Still, if you are the type of person that notices patterns on the landscape, you will have recognized that certain businesses (gas stations, pharmacies, etc.) cluster together in the fashion predicted by this simplistic model, and it’s generally a sign that such businesses are not competing primarily on price, service or quality.

YouTube

Video explanation of the Hotelling Model from TedEd

Huff Model

Another of the measures frequently used by retail site location analysts is the Huff Model. It is used to predict the likelihood of a customer at any distance from a proposed store is likely to shop there. It rests on the premise that people won’t drive very far to shop unless the store is worth the effort. Successful stores, according to the model are bigger, have highly desirable goods and lack competition. Those three elements are all calculated using the formula in the figure at right. To those with math phobias, the formula may look daunting, but it’s nothing more than a series of basic math operations (multiplication, division, exponents, subtraction). Luckily, GIS and spreadsheet programs do most of the work, and software can calculate the formula millions of times over, permitting geographers to estimate the number of customers for many stores simultaneously. Not only could this single formula help a geographic information analyst decide the feasibility of a store (or hospital, etc.), but it also indicates to the marketing department where advertising dollars should be spent; a business geography tactic from the realm of marketing geography.

Figure : Huff Model Formula. See ESRI David Huff