AdClub Digital Day 2024  | Wednesday, May 29th, 2024  | St. Lawrence Centre For the Arts

Case Study: Mitsubishi Motors Canada uses  propensity modelling to increase conversion rate

What is predictive modelling and why should you consider it for your next campaign

Back to All Posts
Blog / AdClub

Case Study: Mitsubishi Motors Canada uses  propensity modelling to increase conversion rate

Case Study 

Mitsubishi Motors Canada uses  propensity modelling to increase conversion rate 

Mitsubishi Motors has pioneered automotive innovation for more than a century.  The global brand established itself in Canada in 2002 and has continued to  champion leading automotive technologies. Its latest car, the Outlander, has virtual  assistance to help drivers with lane centring, speed control, traffic jam stop-and go and much more. With this new vehicle driving Mitsubishi further into the future,  the automaker wanted to channel that innovation into their marketing strategy.

“When you are a cutting-edge brand with a great story to tell,  you need to look for innovative ways to reach consumers“

— Steve Carter, Director of Marketing, Mitsubishi Motor Sales of Canada

The Challenge 

The marketing team at Mitsubishi realized that consumers were increasing their time researching  product information online before purchasing. With this knowledge, the automaker decided to shift to  a digital-first strategy. The “build-and-price” tool on their website took center place in their approach.  This interface allows consumers to compare features and cost before visiting a local dealership.  Mitsubishi used this tool as a “proxy conversion,” because it’s the clearest digital signal of interest  in purchasing the vehicle. The carmaker wanted to acquire more of these “proxy-conversions” which  would eventually help the brand drive more car sales at dealerships.

Mitsubishi’s ad tech solutions partner Dandelion proposed to use Google Marketing Platform and

Google Cloud to harness the power of the brand’s first-party data and predict which consumers were  more likely to use the build-and-price tool. Mitsubishi would then be able to build digital marketing  campaigns to reach those customers.

The Strategy 

Dandelion first exported Mitsubishi’s Google Analytics 360 data to BigQuery, part of Google Cloud.  With BigQuery ML, they created and executed a machine learning model which predicted the likelihood  or propensity of consumers completing the build-and-price tool. This enabled Dandelion to develop a  custom audience that could reach Mitsubishi’s ideal customer.

The next step was to utilize these models. By importing these audiences back into Analytics 360,  Mitsubishi could then make them available for activation across multiple Google advertising solutions.

“Automation is a key component to taking our

digital campaigns to the next level”

— Steve Carter, Director of Marketing, Mitsubishi Motor Sales of Canada

The Conclusion 

With those audiences automatically available for activation  in Display & Video 360, Mitsubishi easily reached their ideal  audience. The brand’s creative digital agency Jan Kelley built  the media campaign using these audiences, determining the  best channels to activate. Over the 6 months that the campaign  lasted, Mitsubishi saw a 81% decrease in their cost per  acquisition and 14x increase in their conversion rate.

After seeing these convincing results, Mitsubishi decided to use  propensity modelling for their future campaigns. They plan to  activate it in other solutions like Search Ads 360 and continue  channelling their innovative spirit.

1.8x decrease in cost-per-action

14x higher conversion rate

70% of total conversions driven by modelled audiences

Dream bigger with data

dandelioninc.ca | info@dandelioninc.ca