By Drew Robb, Contributor
Many companies are embracing big data analytics to unearth sales trends and anticipate opportunities. But General Motors (GM) has taken it a whole lot further.
GM is combining big data, analytics and geographic information systems (GIS) to model dealership performance. Not only that, the company is pushing these capabilities out to its dealers.
This enables dealers around the nation to view local demographics, location characteristics, regional differences and even the competitive brand environment to determine how a given dealership should be performing compared to actual results. This makes it far easier for the company and its salespeople to isolate demand, target its marketing efforts to local preferences and position dealerships to improve success.
“Instead of presenting tabular data or putting dots on maps, we wanted to do real analytics,” says GM’s advanced network analytics manager Bruce Wong. “We need to know things like where our customers live, who buys our cars, and how far they are away from our sites.”
Back in its heyday of the ’50s and ’60s, GM oversaw a vast network of 13,000 dealers. Despite being down to 4,300 dealerships today, the company seeks to use big data to provide that same level of service. It’s been putting the systems in place to achieve that since 1995 and now it is exporting them to the field.
Driving for a deal
Spatial analysis shows many things that would otherwise remain hidden in a vast sea of information. For example, the company now knows that customers will drive two hours to buy a car. Drive-time-based location analytics has isolated the fact that people will drive past the most convenient dealer to save $500 if it’s within that time frame. However, they won’t go nearly so far to get their car serviced. It’s up to GM’s analytics engine to determine where dealerships should be to satisfy the greatest portion of available demand for sales and servicing.
Adding a GIS system known as ArcGIS by ESRI has helped GM successfully add location into the analytics equation. Wong says that it has made a difference in terms of trimming the fat and meeting expectations despite budget downsizing.
“Location analytics helps us to map everything, and do more with less,” says Wong. “We can bring all the data in to it to make better decisions, understand customers and provide better service.”
It is perhaps in the marketing arena that this approach has made the biggest difference. GM budgets around $2 billion each year for marketing. Not so long ago, its strategy was to achieve name recognition via prime time network TV ads, billboards and other traditional channels. But after losing ground to competitors for too long, it sought a different tack.
Instead of hitting the “general public,” it has conducted a series of in-depth analyses to determine the types of households that will buy the various automobiles within its portfolio. It has plotted who and where the buyers are that buy luxury brands, those who prefer midsized and so on for each sales category. By feeding detailed demographic and spatial data to marketing, the automaker can then direct its ad spend to the right areas.
’Chase those households’
“We can chase those households that buy new cars, rather than spending money on households that hold on to older models or only buy used vehicles,” says Wong. “The result is lower spending and higher sales.”
GM also looks to this data when it comes to selecting new locations for dealerships, as well as when to let a particular location go.
“We are using location analytics to question what we did before due to technological and societal change,” says Wong. “More importantly, we have been figuring out how to get that data into the hands of our dealers to help them be more efficient.”
The last 20 years have seen monumental changes in the way business is done and how companies reach customers and potential buyers. Wong notes, for example, that the rise of the Web hasn’t led to many people buying a car on the Internet. Many use it for research and comparison, then head to a dealer.
But that may be about to change. Electric vehicle pioneer Tesla is involved in a legal dispute to be able to bypass dealerships in the sales process. If that happens, the automotive industry might require big data and spatial analytics more than ever to find its customers.
But whatever the future holds, Wong is confident his company is on the right path.
“Location analytics significantly changes your insights into the future,” he says. “You gain an understanding of where you came from and can find out where you want to go.”