Cars.com is using Machine Learning to Predict the Sales of Cars

Pranav Dar Last Updated : 19 Apr, 2018
2 min read

Overview

  • Cars.com has built a machine learning model to help buyers determine when and how to act on a purchase
  • The technology is called “Hot Car” and has been built on over 20 years of data using over 50 factors
  • The initial testing resulted in a double digit increase in sales

 

Introduction

One of the most essential and common tasks of any business is forecasting sales. Before, it used to be done manually by pouring over sheets of numbers. Now, machine learning has taken over and performs the calculations in a matter of minutes (if not seconds).

                                                                                            Source: adage.com

So Cars.com, one of the most popular websites for cars, has announced that it is using machine learning to predict the sales of cars, and to help buyers determine when and how to act on a purchase. This technology is being called “Hot Car” by the company and is already live on their pricing page.

“Hot Car” is being used to predict which vehicles are most likely to sell quickly, based on several variables including car make and model, geographic demand, time on lot, pricing and consumer shopping behaviors. According to Tony Zolla, chief product officer, said that the developers had taken data from the past 20 years and built the model on over 50 factors.

Used vehicles that have a 70% chance of being sold within seven days from their addition to Cars.com will feature a “Hot Car” badge. Brand new vehicles will earn this badge when they had a 70% chance of being sold within 20 days of their addition to the site.

The initial testing phase of using this model resulted is double digit increase in the conversion! It performed particularly strong on mobile.

 

Our take on this

This should be one of the obvious uses of machine learning. It benefits both the user as well as the organization. It crunches the numbers give the buyer insights using which they can make data driven decision on purchasing a car.

This machine learning model will drive up sales, and benefit the dealers by increasing views on car pages, lead conversion and inventory turnover rate. Have you ever used something similar in your projects? Use the comments section below to let us know your views!

 

Subscribe to AVBytes here to get regular data science, machine learning and AI updates in your inbox!

 

Senior Editor at Analytics Vidhya.Data visualization practitioner who loves reading and delving deeper into the data science and machine learning arts. Always looking for new ways to improve processes using ML and AI.

Responses From Readers

Clear

Chinmaya
Chinmaya

Hi Pranav, as part of a consulting assignment for a leading bank in Middle East, we have developed similar approach whereby we have leveraged data from social listening and customer foot prints on the digital platform to identify interests and buying patterns and thereby custom designed credit card products which has resulted in lift in credit card transactions for the bank and at the same time given an opportunity to the bank to upgrade its customers to better card products.

John Mark
John Mark

Hello! I am from USA I saw your blog. It was great, informative and really useful information that you shared with us. I introduced your website to my friends and he was totally impressed keep in touch

John Mark
John Mark

hey It was informative and really useful information that you shared with us.keep wright for us and keep in touch

pebulene
pebulene

Great. I gain something here. Keep posting here informative words.

Congratulations, You Did It!
Well Done on Completing Your Learning Journey. Stay curious and keep exploring!

We use cookies essential for this site to function well. Please click to help us improve its usefulness with additional cookies. Learn about our use of cookies in our Privacy Policy & Cookies Policy.

Show details