Predicting Auto Shopper Behavior
With so much data, stats, reports and metrics, how do we make sense of the most valuable material we have for predicting and preparing for sales in the weeks and months ahead? Predict? Yes! I’m saying that today, by harnessing the power of this data, we can begin to predict, or at least very accurately forecast, what sales will be in the weeks and months ahead, better than ever before. We can observe shopper activity on a dealer level, DMA level, brand level or national level to see shopping trends well before those trends actually start affecting sales. This allows dealers to prepare for what is going to happen instead of being forced to react to what is happening simply because we didn’t have visibility of shopper traffic beyond our own showroom windows.
While watching the weather, I just heard an “expert” explain how a cold front in Oregon is going to affect me here in Nashville in the next few days. How can he tell me how and when an event 3,500 miles away will affect me? And since when could we see a 12-day weather forecast? I remember when they could tell us how hot it would be tomorrow and maybe the next day; beyond that was a mystery. These weather people are getting smarter (and younger) every day. Well, it’s not actually those 20-something talking mannequins’ educated opinions; it’s pure predictive data analysis.
Technology has come so far, and so much data is being produced that we can now see weather 12 days out as a result. We can also see auto shoppers six months out. By aggregating and normalizing auto shopper data, we can see real-time trends in the market at all levels. By tracking four to five major shopper activities, unique visitors, auto shoppers (visitors who actually engage in a vehicle search), total inventory searches and total leads submitted, predictive behavior emerges. As indicated in the graph, we saw a bottom-out of all predictive metrics around September. What happened? Vehicle recalls and bad press. Consumer confidence was at an all-time low and vehicle shoppers fled the market. Hence, September and October sales were down as a result, but the predictor was there.
Look at the lines now. Once the smoke cleared, all major shopping activity was on the rise, dramatically. And we saw better sales in those weeks following the increase in activity. On Black Friday, for example, auto shopping was off 207 percent from a normal Friday. But guess what? Lead submissions were almost unchanged and vehicle searches per shopper nearly doubled. What did we learn? Tire-kickers (or in this case, tire-“clickers”) went to Target while serious auto shoppers took the opportunity to go to the dealerships. If we had only looked at three of the four shopping factors on Black Friday, we would have thought sales would be dead. But in fact, Black Friday weeded out the shoppers, and offered buyers a nice, quiet dealership in which they bought cars.
Moral? Data is the key to running a smarter business in any industry. Auto shoppers allow us plenty of opportunity to observe their behavior and react in plenty of time. In the months ahead, we will explore more ways to run a smart dealership by using data in every facet to predict and prepare for your own future.
Vol. 8, Issue 1