HubSpot Sales Skill

Skill Description: Complete sales workflow management in HubSpot, from cold outreach through close. Covers behavioral analysis, lead scoring, enrichment, personalization, and multi-channel outreach (email + LinkedIn).

When Claude Should Apply This Skill

Apply automatically when:

•Working on HubSpot tasks or contacts
•User asks about lead status, follow-ups, or outreach
•Enriching contact records (/hubspot-enrich)
•Writing personalization content for prospects
•Analyzing engagement signals (opens, clicks, replies)
•Discussing pipeline, deals, or sales activity
•LinkedIn outreach planning

Critical Files (Read Order)

•guardrails.md - MUST READ FIRST. Internal vs external content rules.
•funnel-stages.md - Lifecycle stages and lead status definitions
•scoring.md - Lead scores (custom + HubSpot predictive)
•behavioral-signals.md - How to interpret engagement data
•enrichment-workflow.md - When/how to enrich records
•personalization-rules.md - Writing client-facing content
•linkedin-outreach.md - LinkedIn-specific workflow
•recommended-actions.md - Signal → Action mapping

Core Principles

1. Notes Are Private, Tokens Are Public

Everything you learn about a contact goes in NOTES. Everything the contact might see goes through /brand-brent. See guardrails.md for details.

2. Signals Drive Actions

Don't just report engagement. Recommend specific next actions. "8 clicks" → "Follow up within 48 hours with demo-specific CTA"

3. Enrich Before Personalizing

Always check if enrichment data exists before writing personalization. Run /hubspot-enrich if tech stack, company info, or context is missing.

4. Score-Aware Prioritization

Check both custom scores AND HubSpot predictive scores. High scores + high engagement = immediate attention.

5. Multi-Channel Thinking

Email silence doesn't mean disinterest. Recommend LinkedIn touches. Track which channel gets response.

HubSpot Properties Reference

Lifecycle Stages (in order)

Stage	Meaning
Subscriber	Opted in, no sales activity
Lead	Basic qualification, in outreach
MQL	Marketing Qualified - engagement threshold met
SQL	Sales Qualified - fit + engagement confirmed
Opportunity	Active deal in pipeline
Customer	Closed won
Evangelist	Customer + referral potential

Lead Status (outreach tracking)

Status	Meaning
New	Not yet contacted
Open	Active outreach in progress
Attempted to Contact	Outreach sent, no response
Connected	Response received
Bad Timing	Interested but not now
Past Client	Former customer
Unqualified	Not a fit
No More Emails	Opted out of email

Scoring Properties

Property	Source	Use
`hubspotscore`	Custom	Manual/workflow score
`primary_contact_lead_score_v1`	Custom	Combined fit + engagement
`primary_contact_lead_score_v1_fit`	Custom	Company/role fit score
`primary_contact_lead_score_v1_engagement`	Custom	Activity-based score
`hs_predictivecontactscore_v2`	HubSpot	AI likelihood to close
`hs_predictivescoringtier`	HubSpot	Contact priority tier

Integration with Other Commands

Command	When to Use
`/hubspot-enrich`	Before personalizing, when tech stack unknown
`/brand-brent`	Before writing ANY client-facing content
`/send-email`	After personalization is ready

Quick Decision Tree

code

New HubSpot Task
    │
    ├─► Is contact enriched?
    │       NO → Run /hubspot-enrich first
    │       YES ↓
    │
    ├─► Check engagement signals
    │       • Last activity date
    │       • Email opens/clicks (if visible)
    │       • hs_last_sales_activity_timestamp
    │
    ├─► Check scores
    │       • Custom score
    │       • HubSpot predictive score
    │       • Fit vs engagement balance
    │
    ├─► Determine action (see recommended-actions.md)
    │       • High engagement → Immediate follow-up
    │       • Low engagement → Try different channel
    │       • No engagement → Sequence or drop
    │
    ├─► Write internal insights → NOTES
    │
    └─► Write personalization → Tokens (via /brand-brent)

Data Sources for Enrichment

Source	Data
HubSpot contact	Basic info, activity history, scores
HubSpot company	Industry, size, tech stack (if populated)
Wappalyzer	Tech stack detection
Company website	Products, messaging, differentiators
LinkedIn	Role, tenure, connections, content
HubSpot sequences	Which sequence, which step, engagement

MQL/SQL Definitions (Current State)

Note: MQL and SQL criteria are being refined. Current state:

MQL (Marketing Qualified Lead):

•Engagement threshold met (opens, clicks, form fills)
•Property: mql (enumeration)
•Needs: Clear scoring threshold definition

SQL (Sales Qualified Lead):

•Fit confirmed (right company size, industry, tech stack)
•Engagement confirmed (responded, booked meeting, demo completed)
•Needs: Clear criteria documentation

Action: As we work contacts, document what actually qualifies them. This skill will evolve with real examples.