I begin every weekday by driving through a toll plaza on the Pennsylvania Turnpike to get to work. By this time, I haven’t usually had my morning cup of coffee yet; therefore, my mathematical skills are probably not always up to par. So, I take the easy way out and use my E-ZPass, which saves me the daily burden of counting out change to make my way through the toll booth.

Overall, the E-ZPass system seems relatively straightforward. You use a credit card to open an account and you receive an electronic tag, or transponder, that has your personal billing and vehicle information embedded into it. You put the transponder somewhere on the dashboard or windshield of your vehicle, which then sends a signal to a receiver as you drive through the toll booth that detects your tag, registers your information and charges your account accordingly. When all is said and done, you see the polite green light that says “Thank You” (unless you have a low balance, of course) and you are on your merry way. Quick and simple, right?

Before I began working in market research, I wouldn’t have thought much more about the E-ZPass system other than it gets me to where I need to go quickly. Now that I’m almost a year into my market research career with more of a research-oriented point of view, I got to wondering a little more in depth about the E-ZPass system and how the company conducted its research within the toll-user market to find out if its new toll system would prosper. After a little research, I found that the company used the ever-reliable conjoint analysis method of research.

The scholarly article, Thirty Years of Conjoint Analysis: Reflections and Prospects by Paul Green, Abba Kreiger and Yoram Wind, discusses the use of conjoint analysis in an abundance of studies throughout the past 30 years. One of the studies that this article focuses on is the research done prior to the development and implementation of the E-ZPass system. E-ZPass has been in the works for about 12 years now; the company began its market research in 1992. Two states, New Jersey and New York, had conducted conjoint analysis research using a sample size of about 3,000 to decipher the potential of the system. There were seven attributes used in this conjoint study, such as number of lanes available, tag acquisition, cost, toll prices, invoicing and other potential uses of the transponder. Once the respondents’ data was collected, it was analyzed in total and by region and facility. The study yielded an estimated 49% usage rate, while the actual usage rate seven years later was a close 44%. While both percentages were not extremely high, the company estimated the usage rate would continue to increase in the future.

Green, Kreiger and Wind make a fair point in their article when they say that conjoint analysis has the ability “to lead to actionable findings that provide customer-driven design features and consumer-usage or sales forecasts”. This study serves as a great example to support this statement just by looking at how close the projected usage rate from the data collected ended up being to the actual usage rate. An abundance of the studies that we execute here at TRC use conjoint analysis because of its dependable predictive nature. Whether clients are looking to enter a new product or service into the market, or are looking to improve upon an already existing product or service, conjoint analysis provides them with direction for a successful plan.

In the case of E-ZPass, it’s safe to say that commuters who frequently use toll roads can thank conjoint analysis for its contribution to making those daily drives just a bit easier.