Skip navigation

The Science of Site Selection

In just a few short years, the site selection process for Columbus, Ohio-based Steiner & Associates, a developer of lifestyle and entertainment centers, has undergone a dramatic transformation. Selecting a new location for a high-end retail center used to require company officials to hop into a car with a broker and drive around to look at buildings.

Now, using software from site selection provider GeoVue, Steiner's research department has not only been able to cut the costs of site selection, but also reduce its site selection analysis from four weeks to just a few days.

“It's really changed our decision-making process, and that represents money,” says Matthew Focht, a developer with Steiner.

Beyond the traditional use of site selection tools to save time and money, some companies are even using the software to target potential employees. In addition, the technology enables retailers to evaluate how opening multiple locations in a geographic region will impact the bottom line.

According to Jim Stone, president of Boston-based GeoVue, the software has made several fundamental advances since its introduction in 1993. The software has the ability to simultaneously search for and rate potential sites. One part of the program determines how well a site meets corporate criteria while another evaluates specific candidates for suitability, Stone says. In addition, the software now has the ability to launch in Windows and on the Web.

The costs of site selection technology can vary with the level of sophistication and the amount and kind of data purchased. A standard customized solution that can provide GIS and mapping functions can cost $25,000 in the first year. Some out-of-the-box software programs can be had for as little as $3,000 to $5,000, but often lack the customization and access to demographic data provided by high-end programs.

Demographics on Demand

Sophisticated site-selection software can reveal many details about prospective customers or employees in a potential location. The technology plugs in demographic and geographic data extracted from national databanks and paints a detailed picture (often in the form of a multicolored map) of a site and its surrounding area. For retailers, it can track income levels and potential spending patterns and even make predictions about how much revenue a particular store will generate when placed near particular co-tenants.

“We like to look at many different factors when we assemble data that will create a descriptive tool or a predictive tool,” says Dave Larson, vice president for practice management and analytical services at San Diego, Calif.-based Claritas, a provider of site selection software and services.

Much of the demographic information in these programs is found in databases of federal agencies, such as the U.S. Census Bureau and the Commerce Department, economic development agencies and even local police departments.

From this mountain of data, the software produces printed reports and maps, or scored equations on the selected site. Claritas also can build an analog model that takes into account current locations and factors that predict success, such as co-tenants or the presence of a certain demographic group within the trade area.

Staubach Retail Services' computer mapping and Geographic Information System (GIS) software can produce maps complete with various demographic data. Want to know where two-income families making more than $75,000 can be found in a particular city? The computer can create a four-color map showing the highest concentrations.

“We use the system initially to find the strong markets for retail clients,” says Janie French, vice president of research and technology for Dallas-based Staubach. “A lot of the information we need, you can't just find on a Web site.”

For example, a major theme park developer asked Staubach for average daily temperatures throughout the country. And Barnes & Noble Booksellers needed data on the number of people with college-level degrees in areas company officials were considering for new stores.

Virtual Site Visits

In addition to locating individual demographic groups, software programs can create geographic rings based on distance, or length of drive time. “If I know someone is willing to drive seven minutes to my site, the software can show how many people live within seven minutes,” explains French of Staubach.

Using high-resolution satellite and aerial imagery, elevation data, GPS coordinates and overlay information about cities and businesses, software can now deliver a streaming, 3-D map of the entire globe. One such program is Earthviewer, developed by Mountain View, Calif.-based Keyhole Corp., which enables a user to pan quickly to a particular area and then zoom down to an individual building.

“It's a digital aerial program that allows you to fly from site to site without getting in your car,” says French. “If I'm looking at sites in Dallas, Atlanta and San Francisco, I can be in a client's office with a laptop and go to each site. That gives the client a first-hand look at the site before he physically goes out and sees it.”

Of course, viewing a building via high-resolution satellite imaging is one thing. Actually going out to kick the dirt and see the neighborhood is something quite different. Even the most avid proponents of site selection technology caution that the data alone isn't enough to make a decision.

In Search of Cheap Labor

While retail operations have made the greatest use of site location technologies, the software's value has also become apparent to other industries. An important factor for manufacturers, for example, is proximity to an appropriate labor supply. This can be particularly critical in service businesses as well.

A survey by Dallas-based Trammell Crow Co. found that call centers, for example, spent more than 70% of their annual expenses on labor, while less than 5% was spent on real estate. So, their version of location, location, location is all about labor, labor, labor. And geo-demographic software can help clients find areas with low cost of living (and real estate) and low salaries, thus achieving dual savings.

“With GIS modeling, we're able to input employee types and find where the highest concentrations of those employees can be found throughout the U.S.,” says King White, vice president of the site selection consulting group at Trammell Crow.

A weighted analysis determines which cities have the highest concentrations of employees who fit the desired profile, and how many are employed by competing businesses and industries. Additionally, a labor-rate survey measures salaries for similar jobs in the area.

According to the Trammell Crow survey, labor costs can be shaved by 50%, if corporations relocate away from high-cost metropolitan areas or away from saturated call center environments. An added bonus: these areas typically boast lower training costs — about 5% to 25% annually — due to minimal employee turnover.

Companies are paying attention to these indicators: The Trammell Crow survey found that during 2001 over 44% of new call centers opened in cities with populations under 500,000.

Trammell Crow uses GeoVue software including the iSite mapping and demographic application. Information generated by the program can be incorporated into Microsoft Excel spreadsheets.

Going forward, Stone says the next evolution of the GeoVue software is well under way. The firm is building a product called that can be configured as a Web service and tailored to the needs of many users, from CEOs to field reps.

While these tools can produce powerful and sometimes compelling answers to a company's real estate questions, they are not the final word, by any means, says Brett Bayduss, a senior consultant with Trammell Crow. “Unless they really know what they're looking at — and what each variable means — they're probably not using it effectively,” says Bayduss. “Technology is a great tool, but to be able to really understand it is what makes it an effective tool.”

Randy Southerland is an Atlanta-based writer.

TAGS: News
Hide comments


  • Allowed HTML tags: <em> <strong> <blockquote> <br> <p>

Plain text

  • No HTML tags allowed.
  • Web page addresses and e-mail addresses turn into links automatically.
  • Lines and paragraphs break automatically.