ICP Research Pipeline
9-agent orchestration from tribal knowledge to validated ICP model
Overview
Multi-agent pipeline that interviews stakeholders, extracts ICP signals, validates against CRM data, and produces weighted scoring criteria. The scoring model improves as you close more deals.
Execution Phases
Stakeholder Interview
Structured extraction of tribal knowledge about best-fit customers
Signal Extraction
Identifies firmographic, technographic, and behavioral signals
CRM Validation
Cross-references signals against closed-won deal data
Scoring Model
Produces weighted scoring criteria across 4 dimensions
Persona Generation
Creates buyer personas with pain points and trigger events
Discovery Config
Configures automated prospect discovery using validated criteria
Outputs
- ICP scoring model
- Buyer personas (3-5)
- Discovery criteria
- Enrichment pipeline config
Example
A Series B cybersecurity company engages for ICP development. The 9-agent pipeline interviews 4 stakeholders, analyzes 200 closed-won deals, identifies 4 firmographic and 3 behavioral signals, and produces a scoring model that ranks 500 prospects in under 20 minutes.
Deep Dives
Ready to deploy ICP Research Pipeline?
This agent ships with Knowledge OS. Set up your system in 90 minutes.
Built and maintained by Victor Sowers at STEEPWORKS