FMCG / Sales intelligence
Sales and lead generation
A leading international consumer brand needed to spot commercial opportunities across a large and constantly moving market. The signals were scattered across websites, event listings and other public sources, and finding one was only the start: someone still had to judge whether it fitted the brand, who owned it and what it might be worth.
What was built
The pieces, and how they fit
All of it is configuration on a single instance.
Schedules that trigger collection functions against the chosen sources and APIs
Functions that normalise what comes back and write it into tables on the instance, so the raw material is structured before any model sees it
Named queries exposed to the agent as tools, so it reads market data through reviewed statements rather than generated SQL
An agent that scores each opportunity against the commercial criteria, classifies it and decides which team it belongs to
Schema-constrained output, so scores and categories arrive in a shape the pipeline can rely on
Functions that notify the right team or write the opportunity straight into the CRM through a connector
The result
Sales teams receive structured, prioritised leads instead of a search task. Every run is recorded, so a lead traces back to the source it came from and the reasoning that scored it.
Scheduled functions
Python functions
Database tables
Named queries as tools
AI agents
Schema-constrained output
CRM connectors
Execution tracking
Get started