Customer Analysis powered by AI (Automotive Only for Previous Customers)

What are Percentiles?

Percentiles are a way of comparing a particular value within a larger set of values. Imagine you're at school and you take a test along with 100 other students. Your teacher might say, "You did better than 85% of the class." That means you're in the 85th percentile - you did better than 85 out of 100 students. The closer to 100 your percentile, the higher you are ranked.

Customer Model Application

Now, let's take that concept and apply it to our customer behavior model. Instead of test scores, we're looking at customers' interactions with the advertiser, such as visiting the website, engaging with marketing, and utilizing their vehicle wallet.

After observing and scoring these activities, we rank all the customers based on their scores. The higher a customer's percentile, the more engaged they are and likely closer to making a purchase.

Let's simplify this with our table:

The percentile represents the ranking of the customer based on their interactions compared to all other customers. The UI label indicates the customer's stage in the buying process.

The Heat Bar is a visual way of showing a customer's engagement level. If a customer is at the 0 percentile, they're just beginning their journey with us. We're still trying to create awareness about our products and services, so their Heat Bar is at 10%.

But if a customer is at the 100th percentile, they are showing strong signs of readiness to purchase. These are the customers whose interactions show they are actively engaging with us and are ready to buy. Their Heat Bar is fully lit up at 100%.

This model helps us understand our customers' buying readiness. By seeing where each customer ranks, our AI can better deploy automated household level marketing, and better meet their needs at every stage of the buying journey. It's like having a roadmap that guides our marketing activation with each customer.

Lead Acceleration

Customers can be accelerated by the AI and delivered directly into the CRM.

• Every single customer action and financial/vehicle data point in the platform is an input to drive the AI.
• Each customer is compared to every other customer for that dealership, and uses the above percentiles to score them every day.
• Directly observed actions, behaviors ,and steps taken (i.e. - like seeing customers back on the website, and email open rate) are weighted  the most in the Machine Learning algorithm.

The outcome from the AI acceleration are the customers most likely to be in-market every month based on all the data consumed by the engine.