Common approach
List → enrichment → outreach
Start with a set of companies and look for enough signals to personalize contact.
EMPV / AI LabC
Prototype / commercial intelligence
Opportunity Engine
A commercial research prototype that tries to find companies by starting from observable operational problems instead of starting from lists and enriching them later.
Current milestoneProblem Research → Prospect Research → Qualification.
Experiment
Problem → signals → companies
Research operational patterns first, then identify companies showing those signals and finally try to qualify them with traceable sources.
01 / The workflow
Research, qualification and human control.
The full design moves from Problem Research to Prospect Research, Qualification, Contact Discovery, Outreach Draft, Human Gate and Feedback/Learning. The orchestrator owns state, limits and routing.
02 / What we are testing
Whether problem-first produces materially better opportunities.
The current milestone is not autonomous sales. It is proving that Problem Research → Prospect Research → Qualification can produce a small set of prospects that humans judge better than a generic list.
03 / Why it is interesting
A commercial engine that accumulates method, not just contacts.
Evidence, interpretation and hypotheses stay separate; models are replaceable execution resources and external actions remain behind a human gate.
EMPV / NOTE
Research logic, providers and architecture are still evolving quickly. We do not publish details that would constitute the system's operational advantage.