ICP Refinement Agent
Turn closed-won deal patterns into a living ICP that evolves with your market.
Monthly AI-powered ICP analysis from HubSpot closed-won deals with Apollo.io enrichment. 4-agent AGGREGATE pattern produces 6-dimension ICP report with pattern confidence ratings and period-over-period comparison. Notion ICP report + HubSpot scoring recommendations.
Four Agents. Six Dimensions. A Living ICP That Evolves With Your Market.
Step 1 — Fetcher
Schedule + Code
Schedule Trigger fires monthly (1st 09:00 UTC) or manual Webhook for on-demand runs. Fetcher paginates HubSpot API for closed-won deals over a configurable 90-day lookback window plus a prior 90-day comparison period. Collects associated company and contact data for each deal.
Step 2 — Enricher
Code-only
Augments company profiles via Apollo.io API with firmographics, technographics, and growth signals. Conditional — toggle INCLUDE_APOLLO_ENRICHMENT to skip when Apollo credits are limited. Rate-limited with SplitInBatches and retry backoff.
Step 3 — Analyst
Tier 2 Classification
Sonnet 4.6 receives ONE aggregate call with all enriched deal data. Produces 6-dimension ICP analysis: company_size, industry_vertical, tech_stack, growth_stage, buyer_persona, and deal_dynamics. Each dimension includes pattern confidence ratings (HIGH/MEDIUM/LOW) and period-over-period comparison showing how your ICP is shifting.
Step 4 — Formatter
Tier 2 Classification
Sonnet 4.6 generates a Notion ICP report page with dimension sections, confidence badges, and shift indicators. Also creates a HubSpot task with scoring adjustment recommendations so your lead qualification criteria stay current.
What It Does NOT Do
Does not score individual leads — that is what Inbound Lead Qualifier does
Does not prospect against ICP criteria — that is what Outbound Prospecting Agent does
Does not forecast pipeline revenue — that is what RevOps Forecast Intelligence Agent does
Does not modify HubSpot deal data or contact records — read-only analysis with task output
Does not scrape external websites — all data from HubSpot API and Apollo.io API
Does not analyze lost deals — focuses exclusively on closed-won patterns
The Complete Customer Success Bundle
7 files — workflow JSON, system prompts, TDD, and complete documentation.
Tested. Measured. Documented.
Every metric is ITP-measured. The ICP Refinement Agent turns HubSpot closed-won deal patterns into a living ICP — analyzing 6 dimensions with pattern confidence ratings, period-over-period comparison, and scoring recommendations at $0.28/run.
Agent Architecture
Fetcher (code-only) → Enricher (code-only) → Analyst (Sonnet 4.6) → Formatter (Sonnet 4.6)
Required Credentials
HubSpot (OAuth2), Apollo.io (httpHeaderAuth), Notion (httpHeaderAuth), Anthropic API
Bundle Contents
7
Cost per Run
$0.28/run (avg)
n8n Compatibility
2.7.5
ICP Refinement Agent v1.0.0 — Technical Reference━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━Architecture: 29 n8n nodes, 4 agents (Fetcher → Enricher → Analyst → Formatter)Trigger: Schedule (monthly, 1st 09:00 UTC) or manual WebhookInput: HubSpot API — closed-won deals with company + contact dataEnrichment: Apollo.io API — firmographics, technographics, growth signals (optional)Intelligence: Sonnet 4.6 (Analyst ICP analysis + Formatter report generation)Output: Notion (ICP report page) + HubSpot (scoring task)Cost: $0.28/run (ITP-measured average)ITP: 8 variations, 14/14 milestones PASSBQS: 12/12 PASSTool A: HubSpot (input — closed-won deal data via OAuth2 API)Tool B: Apollo.io (enrichment — firmographics via httpHeaderAuth API)Tool C: Notion (output — ICP report page via httpHeaderAuth API)Intelligence: 6-dimension ICP taxonomy + AGGREGATE pattern (single Analyst call)Cost Value: 0.28
What You'll Need
Platform
n8n 2.7.5+
Est. Monthly API Cost
$0.28/month (monthly runs) + HubSpot/Apollo.io/Notion included tiers
Credentials Required
- ▪Anthropic API
- ▪HubSpot (OAuth2)
- ▪Apollo.io (httpHeaderAuth)
- ▪Notion (httpHeaderAuth, Bearer prefix)
Services
- ▪HubSpot account (OAuth2 with deals, companies, contacts scopes)
- ▪Apollo.io API key
- ▪Notion workspace (integration token with Bearer prefix)
- ▪Anthropic API key
Setup Track
Quick Start
~15 min
All credentials live, n8n running
Full Setup
1–2 hrs
Needs API config + tables
From Scratch
2–4 hrs
No n8n, no credentials
ICP Refinement Agent v1.0.0
$249
one-time purchase
What you get:
- ✓Production-ready 29-node n8n workflow — import and deploy
- ✓Monthly Schedule Trigger (1st 09:00 UTC) or manual Webhook for on-demand runs
- ✓HubSpot API pagination for closed-won deals with company and contact data
- ✓90-day lookback + 90-day prior comparison for trend detection
- ✓6-dimension ICP taxonomy: company_size, industry_vertical, tech_stack, growth_stage, buyer_persona, deal_dynamics
- ✓Pattern confidence ratings: HIGH (>70% deal match), MEDIUM (40–70%), LOW (<40%)
- ✓Period-over-period comparison showing how your ICP is shifting
- ✓Apollo.io firmographic enrichment (conditional — toggle on/off)
- ✓Notion ICP report with dimension sections and confidence badges
- ✓HubSpot task with scoring adjustment recommendations
- ✓AGGREGATE architecture: single Analyst + Formatter calls — $0.28/run regardless of deal count
- ✓Dual Sonnet 4.6: no Opus required
- ✓ITP 8 variations, 14/14 milestones, $0.28/run measured
- ✓All sales final after download
Frequently Asked Questions
How does it differ from Inbound Lead Qualifier?+
Complementary products covering different sides of lead qualification. ILQ scores individual inbound leads against a static ICP definition. ICRA refines the ICP itself from closed-won deal patterns. ILQ tells you if a lead fits your ICP; ICRA tells you what your ICP should be. ICRA's output feeds INTO ILQ's scoring criteria.
What are the six ICP dimensions?+
Company Size — employee ranges, revenue bands, sweet-spot identification. Industry Vertical — dominant verticals, emerging verticals, win rate concentration. Tech Stack — common technologies, technology clusters correlated with wins. Growth Stage — funding stage distribution, growth trajectory patterns. Buyer Persona — champion titles, buying committee patterns, persona shifts. Deal Dynamics — deal size distribution, cycle length, stage progression, discount patterns.
How does pattern confidence work?+
Each dimension is scored by deal match percentage. HIGH confidence (>70% of deals match the pattern) means the dimension is a strong ICP signal. MEDIUM (40–70%) means emerging or mixed signal. LOW (<40%) means the dimension does not reliably predict wins. Confidence ratings help you prioritize which ICP criteria to enforce strictly vs. treat as soft preferences.
Can I run it without Apollo.io?+
Yes. Set INCLUDE_APOLLO_ENRICHMENT to false in the Config Loader. The Analyst still receives HubSpot deal, company, and contact data — enough for company_size, industry_vertical, buyer_persona, and deal_dynamics. Tech stack and growth stage dimensions will have less data without Apollo enrichment.
How often should I run it?+
Monthly is the default (1st of each month at 09:00 UTC). This balances having enough new closed-won deals for meaningful analysis against catching ICP shifts quickly. You can also trigger it manually via Webhook after a big quarter close or whenever you suspect your ICP is changing.
Why is it $0.28/run?+
AGGREGATE architecture. Instead of analyzing deals individually, all deal data is assembled by code-only nodes and sent to the Analyst in a single call. The Formatter also receives one call. Two Sonnet 4.6 calls total regardless of deal count. 12 monthly runs = $3.36/year in LLM costs.
Does it use web scraping?+
No. All data comes from the HubSpot API (deals, companies, contacts) and Apollo.io API (firmographics, technographics). No web_search, no external scraping. This makes the pipeline fast, reliable, and deterministic.
How does it differ from Outbound Prospecting Agent?+
Different purposes using shared data sources. OPA uses Apollo.io to find and prospect against your current ICP criteria. ICRA uses HubSpot closed-won data (optionally enriched by Apollo.io) to refine what those ICP criteria should be. OPA executes your ICP; ICRA evolves it.
Is there a refund policy?+
All sales are final after download. Review the Blueprint Dependency Matrix and prerequisites before purchase. Questions? Contact support@forgeworkflows.com before buying. Full terms at forgeworkflows.com/legal.
Related Blueprints
Inbound Lead Qualifier
Qualify inbound form leads with a 3-agent ILQ scoring pipeline — web research, 4-criteria scoring, and automatic Pipedrive routing.
Outbound Prospecting Agent
Apollo-sourced leads, AI-qualified and personally emailed — zero manual prospecting.
RevOps Forecast Intelligence Agent
AI pulls your entire HubSpot pipeline every week, computes coverage ratio and deal velocity, and delivers a forecast brief with risks, focus areas, and rep leaderboard — to Notion and Slack.