Can Automotive Data Predict Consumer Behavior For Other Verticals?

Travis Meeks Blog

‘Actions speak louder than words.’ The old adage is one that data-driven marketers agree with whole-heartedly. The modern marketer must draw actionable insights from data when using marketing analytics or else lose out in customer acquisition. In short, data must be predictive, precise, and accurate in order to win new business. The relevancy of the data being modeled is crucial for success. After all, different data is predictive of different things, right?

However, the bigger question remains: Can certain behaviors, characteristics, or consumer data be highly correlated with others? Our experience points to an emphatic ‘yes’. Automotive data, specifically the type of auto owned or driven, is one characteristic that is highly predictive for diverse marketing applications.

Our years of analytical expertise have shown that automotive data is predictive of many other behaviors, decisions, and attributes. In our case, the detailed approach that was taken by our team certainly enhanced this data’s performance, and we have seen first-hand how this has come into play when working with clients across many verticals. One area of expertise is the vacation and travel industry. We assist these clients with customer acquisition and retention with our ability to predict travel preferences such as vacation type, international vs. domestic, and how likely an individual is to take a trip in the next year.

For example, one such client in this space was looking to predict consumer cruise line preferences. Our automotive data became a highly correlated and predictive variable for them. We observed that owning a certain car brand is very predictive of the specific cruise you are most likely to take. Who knew a consumer likely to drive a BMW could also indicate that they would love to go on a Viking River Cruise?

CEO Dave Kelly explains why. “The kind of car you drive speaks to the type of person you are. It is a discreetly personal piece of information. We have found that this data acts as a kind of consumer marker. It segments the populations and offers insight into how people spend money.

Because so much thought, feeling, and situational analysis goes into choosing a vehicle, using auto data for analytical modeling yields insight into a customer’s profile and provides a deeper understanding of consumption patterns.”

The car we drive says a lot about us.Alexandra Paul

Unless you are very young, very old, or live in New York City, chances are you have and use a car quite often. In fact, over 90% of Americans drive daily making the car you choose an extremely significant decision. But what drives that decision?

For consumers, different life events, situations, or values demand a certain type of vehicle. Growing families may choose vans or SUVs due to their specific needs. Shouldn’t that same growing family also be interested in family vacations, life insurance, and college savings accounts?

Owning an environmentally friendly vehicle indicates a concern for the Earth. Shouldn’t these drivers have the propensity and income to support ‘green’ causes?

In both cases our research shows strong correlations. Because factors like these are so influential on consumers’ lives, they undoubtedly effect many purchase decisions. This is the kind of insight analytical marketers dream of attaining. Insight like this allows them to identify, target, and ultimately acquire the best prospects for their products.

Even though these examples may seem obvious, assumptions and intuition are not enough in the complex world of marketing analytics. ROI is the only metric that matters for data driven marketers. The great Sherlock Holmes knew “the temptation to form premature theories upon insufficient data is the bane of our profession”. Although the fictional detective was not referring to marketing modeling or predictive analytics, he was on to something. We must allow the data to speak for itself.

Discovering these purchase triggers begins with a good source of auto prospect data.Kelly Idol

At AnalyticsIQ, we have compiled and modeled a vast consumer database using over 120 data sources. Among these sources is known auto-ownership data compiled from the 7 best auto data providers. We took the observed data samples containing information about individuals and the car they own or drive and cloned it to our 210+ million-consumer universe.

Once the data was cloned, our data scientists were able to create individual predictive models for 26 different automotive brands ranging from Subaru to Chevy to Maserati. Due to this level of granularity and detail, we are able to cover approximately 98% of the automotive market. This has allowed us to predict the type of automobile an individual or household most likely drives or owns and whether or not they are in the market for a vehicle with great accuracy. Such forecasts have enhanced customer acquisition and marketing effectiveness for our auto industry clients including a national auto warranty marketer who mails more than 20 million offers annually. While the purpose of these data products is to make predictions around consumers and their automobiles for auto marketing, we see just how diverse the impact of quality auto data can be when used for predictive modeling.

And this isn’t just for offline marketing. All of our powerful data is available for digital marketing campaigns, and our automotive variables are just as predictive in various, digital marketing applications.

So the next time someone tells you that actions speak louder than words, let them know that some actions speak louder than others!