Oct 31, 2025

Lead Scoring with Perpetual ML Suite

Focus your sales efforts where they count. Perpetual ML Suite's Optimum Business Decisioning feature moves beyond simple prediction to directly optimize your business profit from marketing campaigns, demonstrated here through a lead scoring application.

Optimum Business Decisioning: Optimizing Profit, Not Just Accuracy

Traditional Machine Learning (ML) models prioritize statistical measures like accuracy or F1-score. Perpetual ML Suite's Optimum Business Decisioning feature solves this by incorporating your specific financial inputs into the core of the model's decision function.

Crucially, Perpetual ML Suite enables business employees to get state-of-the-art machine learning results without needing to master complex coding or specialized algorithms.

Using the following financial parameters for our lead scoring demonstration:

True Positive (TP)

Input Value: $1000

Financial gain from a successful conversion (Good Lead → Customer).

False Positive (FP)

Input Value: $100

Financial cost of sales time wasted on a dead lead (Bad Lead → Sales Effort).

False Negative (FN)

Input Value: $500

Opportunity cost of a missed good lead (Good Lead → No Sales Effort).

Verified Performance: The Financial Edge

Using the industry-standard Bank Marketing dataset from the UCI repository, we demonstrate the power of Optimum Business Decisioning even against expert methods using the exact same business objective.

A highly skilled machine learning engineer using the advanced combination of LightGBM + CalibratedClassifierCV achieved a total campaign profit of $733,200.

In contrast, Perpetual ML Suite, leveraging our open-source PerpetualBooster algorithm with the Optimum Business Decisioning feature, delivered a total campaign profit of $740,600, a clear demonstration of our superior profit-driven optimization.

This validates that Perpetual ML Suite delivers 1.0% better profit in a marketing campaign compared to the highly skilled machine learning engineer, even when both aim for the same maximum financial outcome.

This increase in profitability is the direct result of integrating your specific cost and benefit variables into our machine learning decision process.