You do the math: measure for measure, site selection process has big-time sales potential - Opinion - restaurant site selection - Column

How many times have you, as a restaurateur, said, "Gee, I think this would be a good spot for my restaurant, but I wish I had an objective measurement of the site's potential"?

Of course, there are a lot of numbers you can look at: residential population, household income, daily car counts, number of local businesses and so on. And all of those numbers are valid indicators of a site's sales potential.

But which ones are the most influential for your restaurant concept? How do they interact and interrelate to define the sales potential for you before a significant investment is made? And how do you factor in the all-important customer lifestyle and dining-behavior patterns of the clientele in the area?

I have developed a process that captures that information in an objective, numerical manner. I call it the site selection modeling process, which involves three distinct steps.

The first step is the development of a customer profile unique to the restaurant concept. The customer profile uses one of the independent lifestyle segmentation systems that categorize the U.S. population into several broad lifestyle clusters made up of a larger number of individual lifestyle segments. The segmentation system then is cross-referenced against customer-dining preferences for specific restaurant industry segments and food types -- i.e., casual dining Asian, midscale steak, quick-service chicken, and so on -- as recorded by various syndicated industry-research studies.

The second step is the development of an ideal site model for the restaurant concept. That model uses the above lifestyle customer profile along with other relevant demographic and geographic variables. A numerical value is determined for each relevant variable based on a projected annual sales volume goal and the size of the concept's primary trading area, for example, a five-mile radius.

The variables then are set up into a primary and a secondary group, based on their degree of relevancy to the specific concept. The ideal site model is constructed in such a manner that the variables, taken together, reflect a trading area that will support the restaurant concept at the stated sales volume.

The third step is the evaluation of specific sites and trading areas. The statistics of each trading area are individually compared with the matching variable in the ideal site model, yielding an index number for each variable. An index of 100 means the specific trading area statistic exactly matches the ideal site model, which in turn means that the trading area will support the required sales volume for the concept on that variable. For example, an index number of 120 would mean the trading is likely to support a volume 20 percent over the projected goal on that variable. Or, stated another way, the trading area is 20 percent more likely to achieve the projected volume goal. Likewise, an index of 72 means the trading area is likely to support a volume 28 percent below the stated goal on that variable.

Individual sites, once their trading area variables are compared with the ideal model, will culminate in an overall index score. That process not only allows each site to be compared against the sales volume goal itself but also allows the sites to be compared with each other to determine which one or ones have the highest probability of sales success.

The site selection modeling process also allows for what I call back-end evaluation in two specific ways. The first way relates to the new sites we have been discussing so far. After the restaurant has been open its second full year, how does the sales volume compare with the model's overall index-score prediction? Once a number of stores have been evaluated that way, how should the primary and secondary variables be weighted in determining their relevancy? Obviously, that type of evaluation requires a long-term approach to the process.

The second way involves a more immediate back-end use of the process. The site selection model will work just as well in evaluating existing restaurant locations. Once the ideal site model is determined, the trading areas of existing restaurants can be matched to it with the same resulting index scoring. Accordingly, the sales of existing locations can be compared with the predictive sales potential of their respective trading area to determine which locations are exceeding, equaling or falling short of their potential.

This method will prove to be a very effective way to allocate your company's resources, including capital, marketing, operational talent and human resources effort. Or, in the case of a single restaurant operation, the model can reveal whether or not you are offering the type of food that the customers in your trading area are interested in eating on a frequent basis.

Mark Katz, a marketing specialist in the restaurant industry with more than 20 years' experience, is an independent consultant in Vernon Hills, Ill. He can be reached at catsco4@msn.com.

COPYRIGHT 2002 Reproduced with permission of the copyright holder. Further reproduction or distribution is prohibited without permission.
COPYRIGHT 2002 Gale Group