Get in touch
Case Studies

Propensity Models

Advanced solution for the strategic selection of points of sale for third-party sales representatives.

Context

In this project, we developed an advanced solution for the strategic selection of points of sale for third-party sales representatives. Using a propensity model based on machine learning, this approach allows task allocation and improves the efficiency of sales actions.

Challenge

With the possibility of third-party sales representatives representing brands on Ambev’s marketplace, it was crucial to identify the points of sale most likely to purchase specific products. The low effectiveness in task allocation for sales representatives resulted in wasted resources and missed opportunities.
Sales Optimization with Machine Learning
The challenge was to structure the problem to fully leverage the latest machine learning methods to identify patterns and predict the likelihood of purchases at each point of sale. An innovative approach was needed to maximize the effectiveness of sales actions.
Approach

Our strategy

Our service is designed to provide a team of dedicated specialists to turn your challenges into opportunities.

1

We used a machine learning-based propensity model to identify the points of sale most likely to buy specific products.

2

We created variables for each global point of sale, capturing relevant data and purchase behaviors, which feed into the model to predict each location’s propensity.

3

We analyzed historical and behavioral data to determine which variables have the greatest impact on purchase decisions, integrating them into the propensity model.

4

We established a continuous evaluation cycle for the model, adjusting it as needed to improve precision and effectiveness.

5

We regularly monitored its performance, comparing predictions with actual results, and adjusted variables and algorithms accordingly.
Results

Measurable Impact and Results

Significant improvement in the effectiveness of task allocation for third-party sales representatives.

More precise allocation, directing sales reps to points of sale most likely to purchase specific products.
Continuous evaluation ensured the model adapted to changes in purchasing patterns, improving its accuracy over time.
A substantial increase in effectiveness rate, rising from 4% to 60%, indicating a much more efficient, data-driven allocation.
Cases

Conheça nossas histórias de sucesso

Turn your data into a competitive advantage.

Contact Oper and discover how our solutions can revolutionize your business.
Talk to Oper