Sales Navigator Browser Command Center

Hard Constraints (Design Around These)

Browser Safety Protocol (NON-NEGOTIABLE)

LinkedIn detects and restricts unusual activity. Every browser operation must:

1. Pace actions: Minimum 2-3 second delay between clicks, 5-7 seconds between page loads
2. Limit session length: Max 45 minutes of continuous Sales Nav activity, then 15-min break
3. Randomize patterns: Don't click through results in perfect sequential order
4. Cap profile views: Max 80-100 per day across all activity
5. Never bulk-scrape: Extract data from 25-50 profiles per session, not hundreds
6. Respect rate signals: If pages load slowly or show warnings, stop immediately and alert user
7. Screenshot evidence: Take screenshots of high-value signals for evidence trail

Identify Request Type

Mode A: Daily Signal Triage (Priority Mode)

This is the 15-minute daily ritual that makes signal-based selling work. Read references/signal-triage-sop.md for the full SOP.

Step 1: Navigate to Alerts Feed

1. Open Sales Navigator homepage
2. Click "All Alerts" tab
3. Filter by alert category (prioritize in this order):
   • Lead career change (job changes - highest signal value)
   • Account growth signals (headcount growth, funding, M&A)
   • Lead shares content (content engagement - conversation permission)
   • Lead viewed your profile (direct attention signal)
   • Senior hires at account (new decision maker signal)

Step 2: For Each Alert - Quick Qualify

Score each alert on Fit + Signal strength:

• Fit check: Is this person/company in ICP? (industry, size, geo, role)
• Signal strength: Is this a Tier 1 signal (job change, funding) or Tier 2 (content share, profile view)?
• Recency: Is the signal <7 days old? (freshness = urgency)

Step 3: Route Qualified Signals

• P1 (Fit + Tier 1 signal): Extract lead data → send to n8n webhook for immediate outbound sequence
• P2 (Fit + Tier 2 signal): Save lead to monitoring list → weekly review
• P3 (Partial fit): Bookmark for context, review next week
• Not qualified: Skip, don't bookmark (avoid 60-day decay clutter)

Step 4: Extract Signal Data

For each P1/P2 signal, capture:

Lead: full name, title, company, LinkedIn URL, profile language
Signal: type, date, detail text (e.g., "moved from X to Y as CTO")
Account: company name, headcount, industry, HQ location
Evidence: screenshot URL or alert text

Normalize to SignalEvent schema (see references/signal-schema.md) for downstream routing.

Mode B: Search & Extract

Step 1: Define Search Parameters

Before running any search, confirm with user:

• Target type: Lead search or Account search?
• Geography: Which countries/regions? (critical for MENA - always split UAE/KSA/Qatar separately)
• Role cluster: What titles/functions? (use Persona if configured, or Boolean)
• Signal filter: Which spotlight/trigger? (Changed jobs, Posted recently, Following company, etc.)
• Exclusions: Who should we filter OUT? (consultants, recruiters, students)

Step 2: Build and Execute Search

Read references/search-recipes.md for pre-built ICP recipes.

Lead Search execution sequence (in Chrome):

1. Navigate to Sales Navigator Lead Search
2. Apply filters one by one (Geography → Seniority → Function → Title Boolean → Spotlight trigger)
3. Count results - if >2,500, add more filters to segment
4. If using Boolean: max 15 operators, ALL CAPS (AND, OR, NOT), straight quotes only
5. Screenshot the search configuration for reproducibility

Account Search execution sequence:

1. Navigate to Sales Navigator Account Search
2. Apply filters (HQ location → Headcount → Growth → Industry → Job opportunities)
3. Count results - if >1,000, segment further
4. Screenshot configuration

Step 3: Extract Lead Data

For each lead in search results:

1. Read: Name, Title, Company, Location, Connection degree
2. Click profile for: Full title, Company size, Profile language, Recent activity
3. Note any spotlight badges (Changed jobs, Posted recently)
4. Save to appropriate list in Sales Nav

Extraction pacing: 3-5 seconds per profile view, max 25-50 per session.

Step 4: Output

Compile extracted data in structured format for downstream:

• If routing to Clay: Name, LinkedIn URL, Company, Title, Signal type
• If routing to HeyReach: LinkedIn URL (trailing / required), First name, Company, Position
• If routing to GHL: Name, Email (if visible), Company, Title, Signal tags

Mode C: Search Engineering (Weekly Refresh)

Read references/saved-search-strategy.md for the full strategy.

Monitoring Stack Design

Instead of 50 narrow saved searches, build "monitoring stacks" - each combining Fit filters + one signal filter:

Weekly Refresh Checklist

1. Review saved search results count - any that produce 0 weekly? Remove or broaden
2. Review saved search results quality - too many false positives? Add exclusions
3. Check if ICP shifted - update Personas if needed (max 5)
4. Add accounts from pipeline focus to saved account lists
5. Prune lists that exceeded 1,000 leads (move cold to archive list)

Mode D: List Management

List Architecture

Organize lists by purpose, not org chart:

List Hygiene Rules

• Max 1,000 leads per list (hard limit)
• Review and prune monthly
• Move "contacted" leads to Pipeline lists
• Move "no response after sequence" to Archive
• Never delete leads entirely - archive for future re-activation

Mode E: Account Intelligence Deep Dive

For priority accounts, run a structured deep dive:

Step 1: Account Page Analysis

1. Navigate to account page in Sales Nav
2. Capture: Headcount, Growth %, Industry, HQ, Recent activities
3. Check "Growth Alerts" and "Risk Alerts" sections
4. Note any recent news or updates

Step 2: Map Buying Committee

Use Relationship Explorer (available on Core for account pages):

1. Navigate to "People" tab on account page
2. Filter by your Persona (or manually by seniority + function)
3. Identify the 3 roles: Economic Buyer (CXO/VP), Champion (Director/Manager), Gatekeeper (IT/Procurement)
4. Check connection degree for each - 1st degree = warm path exists

Step 3: Output Account Brief

Account: [Company Name]
HQ: [Location] | Headcount: [X] | Growth: [+/- %]
Industry: [X] | Recent Signals: [list]

Buying Committee:
- Economic Buyer: [Name, Title] - [connection degree]
- Champion: [Name, Title] - [connection degree]
- Gatekeeper: [Name, Title] - [connection degree]

Recommended Entry Point: [who to contact first and why]
Signal-Based Angle: [which signal to reference in outreach]

Mode F: Setup & Configuration

Persona Configuration

Sales Nav allows 5 personas (2 defaults + 3 custom):

1. Assess user's ICP - what are the primary buyer roles?
2. Navigate to Personas settings
3. Configure each persona: Function + Seniority + Title keywords + Geography
4. Test each persona in lead search to validate result quality

MENA tip: Build separate personas for English-title and Arabic-title roles if your ICP spans both language patterns.

Initial Saved Search Setup

Help user design their first set of monitoring stacks using the framework in Mode C.

MENA-Specific Playbook

Read references/mena-playbook.md for the full MENA guide. Key rules:

• Profile language filter: Pin this as default. Split searches into English and Arabic runs
• Geography: Always use "Headquarters location" for account searches + "Lead geography" for lead searches. Many MENA execs HQ in London/Singapore
• Business hours: Sunday-Thursday. Half day Thursday. No Friday outreach
• Ramadan 2026 (Feb 28 - Mar 30): Reduce volume 50%+, shift to evening timing
• Bilingual Boolean: Build parallel title strings: ("Sales Director" OR "مدير مبيعات")
• Dual role profiles: Higher false positives in MENA due to title localization. Always add seniority filter as guardrail

Signal-to-Action Routing

Self-Learning Protocol

After each session, capture:

1. Which searches produced highest quality results (filter combos that work)
2. Which alert types generated the most actionable signals
3. False positive patterns to exclude next time
4. Any LinkedIn UI changes or new features observed
5. MENA-specific discoveries (title patterns, geo quirks, language patterns)

Update references/learnings-log.md with dated entries.

Coordination with Other Skills

Reference Files