Several Methods, Still a Challenge
By Robert Roman
To help bring your car wash business idea to fruition, location is one of the first challenges to address. Solving this problem is even more important today because the economic and financial environment has created a challenging development outlook.
A car wash works best when it serves the needs of the community — locating one in a commercially attractive setting is one of the keys to success. Consequently, having a sound strategy is one of the keys to solving the location problem.
One method is to imitate or mimic another retailer. Consider a mass merchandiser that decides a market zone has the attributes to justify building one of the company’s big-box, wholesale retail outlets. Subsequently, available space surrounding this new store invariably becomes populated with mainstream franchises and special-uses like gasoline, convenience store, dry cleaner, auto parts, etc., which are thought to serve as good attractors for a car wash.
Another approach is to rely on the judgment of experts. This approach is based on the assumption that an experienced person can make a good decision by applying his or her knowledge to the problem. In applying this technique, the person either develops judgmentally his/her own opinion (i.e., poor/fair/excellent, low/med/high) or obtains opinions from other experienced people. The opinions are then combined into a single one.
Another way to solve the location problem is to use a systematic approach. This involves specifying market zone by size (i.e. radial, drive time), taking market potential into consideration, and applying analytical models.
For example, analog models are often used to rank market zones based on desirable attributes such as population, income, lifestyle, growth vectors, traffic flow, business counts, competition, etc. Analog models are based on the principle of similarity where the most likely performance of a retail outlet will be the average of some subset that has the same characteristics as the subject.
Cause and effect models like traffic count, capture rate, or gravity models are widely used methods to predict car wash sales volumes. The capture rate method is based on the assumption that there is a strong, positive relationship between car wash volumes and the amount of traffic that passes by a site.
Gravity models are based on the assumption that the attraction of consumers to a given retailer is directly
proportional to the quality of the retailer and inversely proportional to the distance of the retailer. Gravity models predict sales at specific retail sites based on relevant aspects of store performance such as accessibility, appeal, price, brand, operations, and location. It measures these things in terms of all adjacent sites that offer any competition to the subject.
Any one of the approaches described above can be used to help solve the car wash location problem but each one has limitations that may affect the outcome.
Imitating the location strategy of another retailer ignores the fact that the characteristics of markets also depend on the type of business and competitors’ locations and characteristics. The weakness of the judgmental approach is the need for knowledge of probability theory plus insight to the car wash business on the part of the person(s) developing the opinion. There are also issues that can be encountered with the systematic approach.
Analog models ignore the fact that consumers within a market zone can migrate outside the zone just as consumers outside the zone can migrate to it. Common experience shows that the relationship between car wash volumes and traffic is not strong or statistically significant. Gravity models involve a high degree of subjectivity in rating the relevant aspects of store performance. They tend to work best for retail outlets that sell commodities and services that have broad appeal and market penetration, whereas car washing is a nonessential service that is nice to have but it is not necessary to survive.
One way to overcome these limitations is to combine different outcomes into one good one. There are several benefits to this. If one method produces a result that is significantly different from another method, for example, you can be certain that at least one of them is grossly wrong. No matter how well a particular method has worked in the past, at one time or another it will fail due to unforeseen circumstances that have an impact on the outcome.
Merging outcomes can be accomplished judgmentally by taking the straight average or by statistically computing a weighted average or an expected outcome (i.e. beta probability distribution), which is a function of the optimistic, most likely, and pessimistic outcomes.
Common experience has shown that solving complicated problems like retail location and anticipated sales does not occur in a vacuum. With this in mind, two or more intelligent heads are often better than one.
Bob Roman is president of RJR Enterprises — Consulting Services (www.carwashplan.com) and vice president of Bubble Wash Buildings LLC. You can reach Bob via e-mail at firstname.lastname@example.org.