Next best offer - What is the ideal price a customer should be offered to buy a car?

changing discounts
Data science

The challenge

It is common practice during the sales process to give customers discounts to encourage the sale, but what should be the ideal price? One strategical approach of the company to increase its profits was to optimize this question. Previously, dealers requested a custom discount on the price manually from the head office. This strategy was labor intensive, and ran the risk of allowing discounts that were higher than necessary as they were based on gut instincts and not data-driven. The challenge was to create an intelligent pricing recommendation tool which automatically balances the discount (as low as possible) against the sales probability (as high as possible) for each car, based on constraints defined by the business.

Our solution

Contiamo created an intelligent solution which consisted of two stacked components, a machine learning model and a discount-sales-probability optimizer. The developed data science pipeline processes daily data deliveries into snapshots. The on a daily base produced output, pricing recommendations, are then entered into the customers’ systems. Creating an explainable, transparent model was especially important for this project to ensure the adaption by the dealers.

Tech stack


The project

Proof of concept

The first steps consisted of data exploration and collaborative definition of aspects to be considered, such as business objectives, ethical issues in handling demographic data, modeling trade-offs between explainability and predictability.

During the modeling, we identified a stacked 2-component solution as the ideal solution to recommend prices and discounts. First, sales probabilities for each car were predicted using a gradient boosted tree model. In the second step, we identified price-elasticity curves as a suitable and efficient way to display how the probability of a sale depends on the discount offered. The ideal discounts are selected based on these insights and optimization rules.

We included model output explanations (SHAP) in our price recommendations. This  was the key to increased confidence in the model and allowed business users to review the recommendations for specific cars when needed.

Roll out

For the initial roll-out, we provided discount recommendations in daily batches, with data exchanged via a secure file-sharing system. After a successful initial roll-out, we helped the customer to host the pipeline in their infrastructure thanks to our flexible docker-based framework.

Later, the customer carried out a complete redesign and fundamental renewal of its data environment. The pipeline was transferred from the legacy environment to the new cloud platform accordingly. Again, a Docker-based solution enabled a flexible and easy transfer to the newly created DevOps pipeline.

Improvement Phase

Our constant monitoring helped to detect data integrity issues in the daily data deliveries as well as edge cases in the model. We further assisted the client in redefining the strategy for rebate components in their systems, which impacted both the input and output of the model.

Achieved benefits

tick mark

Optimized discounts for > 30,000 requests on cars

tick mark

Sales success increased by 25%

tick mark


The goal was to increase the customer’s profit by maximizing the sales success while minimizing offered discounts. Contiamo created a pipeline ingesting daily data deliveries and applying two stacked algorithms to generate price recommendations for all cars in stock. The solution was successfully implemented in the market and optimized daily discounts for more than 30,000 cars over 3 years. It will be transferred to further markets. By including output explanations in our discount recommendations, we increased business confidence and allowed the customer to review the recommendations for specific cars when needed.

Are you interested in more information? Let's talk about your current topics and challenges!