← Back to Home
AI-Driven B2B Collections Optimization
Major Telecom Enterprise • Process Intelligence • Tech & Ops Lead
Context & Challenge
A major telecom enterprise faced significant inefficiencies in their B2B collections process. High volumes of accounts were managed manually, leading to prioritization issues, lack of visibility on customer promises, and inconsistent follow-ups. The operation relied on scattered data across legacy systems, making “reactive” dialing the only option.
My Role
Acting as the Process Translator & Tech Liaison, I bridged the gap between the Operations team and Data Science/Dev units. I led the Discovery phase, defined the Proof of Concept (PoC) scope, and validated the GenAI outputs against real operational needs to ensure the tool was usable, not just theoretical.
The Approach
“This project reflects my core focus: translating operational problems into practical, data-informed decision systems.”
We moved from a manual “reactive” model to a data-driven “proactive” flow. The system ingests call transcripts, scores the interaction based on customer intent, and automatically prioritizes the next action for the agent.
How prioritization decisions were supported during the pilot.
Operational Insight Examples (Real Data)
Examples of how conversational data was transformed into actionable operational signals during pilots.
*Dashboards shown as originally used in enterprise environments (Spanish), reflecting real operational contexts.