Resale price prediction - vehicle sales data - CRM-integration - machine learning - automotive marketplace
One of our customers operates a large vehicle resales platform in the Netherlands, serving as an intermediary between private sellers and auto garages. By offering a transparent online service, the platform allows individuals to easily submit their car’s license plate and, optionally, a desired sales price. The make and model are automatically identified, after which an account manager contacts the owner to confirm the asking price and gather additional details such as features, condition, and any damages. The car, along with photos, is then listed for auction to a network of vehicle garages, with the highest bid winning and the platform taking a commission on the sale.
To support a smoother process for both sellers and the company, there is significant value in accurately estimating expected resale prices early on—both for informing potential sellers if their price expectation is realistic, and for helping to match cars with interested buyers more efficiently. The platform’s sales history, including bid prices and detailed information on previously sold vehicles, provides a rich source of data for building such predictive tools.
For this assignment, you will develop a statistical or machine learning model to predict the expected resale value of vehicles using historical data. You will analyze factors such as make, model, year, condition, reported damages, extra features, and previous auction results to train your model for accurate price estimation. Your deliverable will be a working prediction tool that can be integrated into the current CRM application, helping account managers quickly assess vehicle value at the start of the sales process.
Get in touch with Martijn, founder of Bullit.
+31 6 39 56 09 34