Directions:

Copy and paste the prompt below into your company's authorized Gen AI too to determine Lead Scoring.

# B2B Lead Scoring Analysis Prompt
## Input Data Requirements

Please ensure you have the following CSV files with their required fields:

1. leads.csv
   - lead_id
   - company_name
   - industry
   - annual_revenue
   - employee_count 
   - country
   - state
   - lead_source
   - lead_status
   - created_date
   - last_activity_date
   - assigned_rep_id

2. quotes.csv
   - quote_id
   - lead_id
   - quote_date
   - total_amount
   - product_category
   - quote_status
   - close_date
   - probability
   - competition
   - loss_reason
   - modified_date
   - sales_rep_id

3. quote_items.csv
   - quote_id
   - item_id
   - product_id
   - product_name
   - category
   - quantity
   - unit_price
   - discount_percent
   - margin_percent
   - line_total

4. activities.csv
   - activity_id
   - lead_id
   - quote_id
   - activity_type
   - activity_date
   - activity_result
   - next_step
   - notes
   - sales_rep_id

## Analysis Instructions

Using the provided CRM data files, please:

1. Calculate lead scores based on the following weighted criteria:
   
   A. Engagement Score (30%)
   - Meeting/demo attendance (40%)
   - Email/call response rate (30%)
   - Quote request frequency (20%)
   - Website/document interactions (10%)

   B. Company Profile Score (40%)
   - Annual revenue match (35%)
   - Industry alignment (25%)
   - Employee count fit (20%)
   - Geographic location (20%)

   C. Behavioral Score (30%)
   - Quote history (40%)
   - Response time to quotes (25%)
   - Purchase timeline clarity (20%)
   - Budget authority (15%)

2. Generate lead grades using these score ranges:
   - A: 90-100 (High potential)
   - B: 70-89 (Strong prospect)
   - C: 50-69 (Moderate potential)
   - D: 30-49 (Needs nurturing)
   - E: 0-29 (Low potential)

3. Create a dashboard visualization showing:
   - Lead score distribution
   - Key performance metrics
   - Industry performance comparison
   - Lead source effectiveness
   - Sales rep performance

## Desired Outputs

1. Dashboard Components:
   - Lead grade distribution chart
   - Conversion metrics by score grade
   - Industry performance comparison
   - Lead source distribution
   - Sales activity impact analysis

2. Key Metrics Display:
   - Quote-to-win rate
   - Average deal size
   - Sales cycle length
   - Overall engagement score

3. Strategic Insights:
   - Top-performing industries
   - Most effective lead sources
   - Best-converting sales activities
   - Risk factors in low-scoring leads

## Analysis Parameters
- Analysis timeframe: Rolling 12 months
- Minimum activity threshold: 2 interactions
- Score calculation frequency: Daily
- Conversion rate baseline: Industry average
- Activity recency weight: Last 90 days prioritized

## Example Implementation

```python
# Sample code structure for lead scoring
import pandas as pd
from datetime import datetime, timedelta

def calculate_engagement_score(activities_df, lead_id):
    """Calculate engagement score based on activities"""
    lead_activities = activities_df[activities_df['lead_id'] == lead_id]
    
    # Weight different activity types
    activity_weights = {
        'Meeting': 1.0,
        'Demo': 0.8,
        'Call': 0.6,
        'Email': 0.4
    }
    
    # Calculate weighted score
    activity_score = sum(activity_weights.get(act_type, 0.5) 
                        for act_type in lead_activities['activity_type'])
    return min(100, activity_score)

def calculate_company_score(lead_data):
    """Calculate company profile score"""
    revenue_score = revenue_scoring_logic(lead_data['annual_revenue'])
    industry_score = industry_scoring_logic(lead_data['industry'])
    employee_score = employee_scoring_logic(lead_data['employee_count'])
    
    return (revenue_score * 0.35 + 
            industry_score * 0.25 + 
            employee_score * 0.20 + 
            location_score * 0.20)

def calculate_behavioral_score(quotes_df, lead_id):
    """Calculate behavioral score based on quote history"""
    lead_quotes = quotes_df[quotes_df['lead_id'] == lead_id]
    
    quote_score = quote_scoring_logic(lead_quotes)
    response_score = response_time_scoring_logic(lead_quotes)
    timeline_score = timeline_scoring_logic(lead_quotes)
    
    return (quote_score * 0.40 + 
            response_score * 0.25 + 
            timeline_score * 0.35)
```

## Data Processing Notes
- Handle missing values appropriately
- Normalize scores to 0-100 scale
- Account for industry-specific benchmarks
- Consider seasonal patterns
- Weight recent activities more heavily

## Visualization Guidelines
- Use consistent color scheme
- Include interactive tooltips
- Show trend lines where applicable
- Highlight significant deviations
- Enable filtering by time period

## Recommended Actions
1. For A-grade leads:
   - Immediate sales follow-up
   - Executive engagement
   - Custom solution development
   - Priority quote processing

2. For B-grade leads:
   - Regular sales contact
   - Product demonstrations
   - Case study sharing
   - Competitive analysis

3. For C-grade leads:
   - Nurture campaigns
   - Educational content
   - Needs assessment
   - Value proposition reinforcement

4. For D-grade leads:
   - Lead nurturing
   - Market education
   - Requirement clarification
   - Budget discussion

5. For E-grade leads:
   - Marketing automation
   - Periodic review
   - Resource optimization
   - Segment reevaluation