Market intelligence with lead scoring

Designed and implemented a machine learning classification model to score and rank companies within the active customer base. By combining data-driven behavioral patterns with strategic human inputs, the model defined "target companies" to align field sales resources and marketing efforts for conversions

The sales team had a large active customer base with no systematic method for prioritizing outreach. Account selection relied on individual intuition, causing high-potential accounts to be overlooked and low-potential ones to consume disproportionate time, leading to missed opportunities and inefficient resource allocation.

Integrated data from sales activity logs, purchase history, geographic variables and human input to build a feature-rich dataset. The model featured a mechanism allowing us to intentionally weight metrics based on sales leaders command, aligning the algorithm with their specific business strategies A supervised classification model was trained to score each account based on that and the output was a ranked list focused on the accounts with the highest conversion and upsell potential.

Systematic, data-driven account prioritization replacing ad-hoc intuition. Improved alignment between field deployment and revenue opportunity. The model was embedded into the team's regular planning cycle, creating a feedback loop for continuous refinement.