Lead Research Skill

Research LinkedIn prospects to gather personalization opportunities, ice breakers, and conversation starters for more effective outreach.

Process

Step 0: Load Defaults

First, check for saved decisions:

Read .business_growth/sales/DECISIONS.md if it exists. This file stores:

Sales Navigator availability
Buyer persona and problem we solve (useful for framing research)
Target industries (helps identify relevant talking points)
Competitors to avoid mentioning

If DECISIONS.md exists and has Sales Navigator info:

Use the saved preference, no need to ask again
Reference the buyer persona to focus research on relevant pain points

If DECISIONS.md doesn't exist or missing Sales Navigator: Ask: "Do you have LinkedIn Sales Navigator? It provides deeper profile insights."

Why Sales Navigator matters:

With Sales Navigator: Access full profile info, see who viewed them, more activity history, notes/tags
Without Sales Navigator: Standard profile view, may have limited visibility for non-connections

Offer to save: If user answers and no DECISIONS.md exists, offer to save this preference.

Step 1: Identify Target Lead

If linkedin_url provided:

Use directly for research

If lead_name provided:

Search in campaigns under .business_growth/sales/campaigns/
Find matching lead and get LinkedIn URL

If campaign_id provided (without specific lead):

Load LIST.md from .business_growth/sales/campaigns/<campaign_id>/
Ask user which lead to research, or research all leads

If neither provided:

Check for existing campaigns in .business_growth/sales/campaigns/
If campaigns exist, ask user:
- (A) Create new campaign → Generate a meaningful snake_case name
- (B) Add to existing campaign → Select from available campaigns
Establish campaign_id before proceeding
Ask user for LinkedIn URL or lead name to research

Step 2: Navigate to LinkedIn Profile

Get browser context with tabs_context_mcp
Navigate to profile:
- With Sales Navigator: Use linkedin.com/sales/lead/<id> format for richer data
- Without Sales Navigator: Use standard linkedin.com/in/<username> URL
Take screenshot to verify profile loaded
Check for any access restrictions (may need to be connected for full profile)

Step 3: Extract Profile Information

Use read_page to extract:

Basic Info:

Full name
Current title and company
Location
Profile headline
Connection degree

About Section:

Summary/bio content
Key themes and interests
Communication style

Experience:

Current role details
Previous companies
Career progression
Time in current role

Education:

Schools attended
Degrees and fields
Graduation years

Skills & Endorsements:

Top skills
Endorsement counts

Step 4: Analyze Recent Activity

Scroll to activity section and extract:

Recent Posts (last 30 days):

Topics they post about
Engagement levels
Opinions and perspectives

Reactions & Comments:

Content they engage with
Thought leaders they follow
Interests indicated

Articles Written:

Published content
Expertise areas

Step 5: Identify Personalization Opportunities

Analyze gathered data for:

Conversation Starters:

Recent post to reference
Shared connections
Common background (school, company)
Mutual interests

Pain Point Signals:

Job changes (new role = open to tools)
Company growth signals
Hiring activity
Complaints or challenges mentioned

Rapport Builders:

Hobbies and interests
Causes they support
Content preferences

Timing Signals:

Best time to reach out
Upcoming events
Recent achievements to congratulate

Step 6: Generate Ice Breakers

Create 3-5 personalized ice breakers based on research:

Types of Ice Breakers:

Content Reference: "Loved your post about X..."
Shared Connection: "I see we both know Y..."
Common Background: "Fellow [school/company] alum here..."
Industry Insight: "Given your focus on X, thought you'd find..."
Congratulations: "Congrats on the recent [achievement]..."

Step 7: Save Research

Save to the campaign folder: .business_growth/sales/campaigns/<campaign_id>/leads/<lead_name>/research.md

Lead Naming: Generate meaningful snake_case names for leads based on their name and company (e.g., john_smith_acme, sarah_jones_techcorp). Keep names concise, lowercase with underscores, no special characters.

Create research file:

# Lead Research: <Full Name>

## Research Metadata
- **Researched**: <ISO timestamp>
- **LinkedIn**: <url>
- **Campaign**: <campaign_id if applicable>

## Profile Overview

### Current Position
- **Title**: <title>
- **Company**: <company>
- **Duration**: <time in role>
- **Location**: <location>

### Background
- **Previous Role**: <most recent previous>
- **Education**: <school, degree>
- **Career Theme**: <progression summary>

### About
<summary of their bio/about section>

## Recent Activity (Last 30 Days)

### Posts
1. **<date>**: <topic/summary>
   - Engagement: <likes/comments>
   - Key quote: "<relevant excerpt>"

2. **<date>**: <topic/summary>
   - Engagement: <likes/comments>

### Engagement Patterns
- Topics they engage with: <list>
- Thought leaders followed: <names>
- Activity level: Low / Medium / High

## Personalization Opportunities

### Strong Signals
1. <signal with context>
2. <signal with context>

### Potential Pain Points
1. <inferred pain point>
2. <inferred pain point>

### Shared Context
- Connections: <mutual connections>
- Background: <shared experiences>
- Interests: <common interests>

## Recommended Ice Breakers

### Option 1: Content Reference
"<personalized message referencing their recent post>"

**Why this works**: <explanation>

### Option 2: Industry Insight
"<message offering relevant insight>"

**Why this works**: <explanation>

### Option 3: Mutual Connection
"<message leveraging shared connection or background>"

**Why this works**: <explanation>

## Outreach Recommendations

### Best Approach
<recommended outreach strategy>

### Timing
<best time/day to reach out>

### Tone
<recommended communication style based on their profile>

### Topics to Avoid
<anything that might not resonate>

## Raw Notes
<any additional observations>

Step 8: Report Summary

Provide user with:

Key findings summary
Top 2-3 personalization angles
Recommended ice breaker
Suggested next steps

Browser Tools Used

Tool	Purpose
`tabs_context_mcp`	Get browser context
`navigate`	Go to LinkedIn profile
`read_page`	Extract profile content
`computer`	Scroll to load more content, screenshot

Research Quality Tips

Good Personalization

Specific and recent (last 30 days)
Shows genuine interest
Relevant to outreach purpose
Not creepy or over-researched

Avoid

Mentioning personal/family info
Referencing very old content
Being too detailed (feels stalky)
Generic compliments
Assumptions about their problems

Integration with Campaign Execution

When research is saved to a campaign lead folder:

{{custom}} token in templates can reference research findings
Campaign execution skill will use research for personalization
Higher quality outreach leads to better response rates

lead-research