STI: WorkPlace - Methodology

Every market-focused company uses demographic data to target consumers where they live. But only the most strategic companies also target consumers where they work using workplace estimates. Why do some companies care about workplace estimates? Because consumers who spend a large percent of their waking hours working in an area that is different from where they live, also purchase products and services near their workplaces from restaurants, banks, dry cleaners, drug stores, and more. So, by determining the workplace populations of markets, companies gain much more precise market insight and, therefore, more profitable site selection capabilities.

However, companies that are not using workplace population estimates are not ignoring them only because they lack an understanding of their power to support better market research, but also because the traditional workplace estimates have not delivered accurate numbers. STI: Workplace solves this long-standing problem by bringing accuracy to estimating workplace populations through an innovative new methodology. In fact, by using new workplace data sources and an innovative "bottom-up" analytic methodology, STI: Workplace delivers the most accurate workplace estimates available today.

What's more, STI: Workplace adds considerable value to workplace market estimates by providing not just how many people work in a particular industry, but also their occupations and income levels. This additional consumer insight gives competitive, profit-focused companies a tremendous strategic advantage when deciding which location is the most likely to attract a large number of consumers and, ultimately, net the maximum return on investment.

The Traditional Approach to Workplace Estimates

The traditional workplace estimating tools that have been available up until today all share one significant fundamental problem: They are based on business data from database providers whose primary business is selling marketing lists to list brokers - not creating data for statistical application. To obtain their business data, these firms employ large staffs of telephone solicitors who call companies listed in business telephone books in each market and ask them as series of basic questions, including how many people work in their companies. While at first glance, this process appears to be an accurate way to determine workplace population estimates, a closer examination reveals a serious flaw.

In fact, these data firms are frequently double-counting the number of people who work in companies. How? By calling many of the companies more than one time if their telephone numbers are listed more than once. The data firms do not maintain adequate controls for ensuring that the telephone solicitors count each business only one time.

Knowing this, it's easy to understand how off-target workplace estimates based on these employee counts can be. As further evidence of the extent of the problem, in 2004 one such data provider estimated that there were 12.5 businesses in the U.S. However, three respected U.S. agencies reported that there are 7.5 million businesses in the U.S. This represents an error of five million businesses.

Furthermore, while three respected U.S. agencies reported that there are about 7,000 businesses in the U.S. with more than 1,000 employees in a single location, the data firm stated that there are more than 10,000 business with more than 10,000 employees at a single location. With such vast discrepancies in employee and company counts, it's not surprising that more businesses are not using traditional workplace estimates in their site selection decision-making process.

An Innovative Approach to Workplace Estimates

STI: Workplace does not use workplace estimates from the traditional database providers, as most other data product providers do. Instead, our product's data is derived from three respected U.S. agencies:

• U.S. Census Bureau
• U.S. Bureau of Labor Statistics
• U.S. Postal Service

For each market, the business unit estimates from these three agencies are run through a sophisticated mathematical process that captures the data where the three databases correlate. The end result is a single workplace estimate for each market. We then expand that estimate to a current time frame by using zip-plus-four data and historical averages at the market level. In other words, we do not apply national esti-mates to the calculation, but instead make the data more relevant for market research by taking it down to the local level. (Note: The methodology for STI: Workplace is similar to STI: PopStats. For a fuller expla-nation of the bottom-up approach with zip-plus-four data, please refer to the STI: PopStats methodology document.)

Our second step, after determining the number of businesses in a market, is to determine the number of employees by assessing the historical averages that are typical within each individual market. So that, for example, when estimating the number of employees for a typical law firm in Austin, Texas, we use the historically averages for law firms in Austin, Texas.

The third step is to determine what the occupations of the employees are. Again we refer to the historical data and relationships to determine typical job breakdowns per business per employees at the market level.

Our fourth and final step is to determine an income estimate associated with each occupation in each workplace, again referencing the market level historical averages. This income-estimating information is unique from every other workplace population product available. It is especially valuable for companies that have found that employees with particular income levels are attracted to their products and services.