Scoreboard - AI- Powered Lead Scoring

This article explains the data fields included in the BankingBridge lead scoring dataset and how they help identify and prioritize high-intent leads.

1. Lead Identification

• id: Unique identifier for the lead scoring record
• subscription_id: Links the score to a specific lead/subscription in the system

2. Lead Score & Classification

• intent_score: Numeric score (0–100) representing the lead’s likelihood to convert. A score over 80 is “Hot”.
• category: Classification of the lead based on engagement and intent
  • Example: hot_lead, warm_lead, cold_lead

3. AI Summary & Reasoning

• summary: A concise, human-readable overview of the lead’s intent and behavior
• reasoning: Detailed explanation of why the lead received their score
  • Includes behavioral signals
  • Timeline of engagement
  • Financial strength indicators
  • Risk factors (if any)

This section provides transparency into how the AI evaluates each lead.

4. Recommended Action

• suggested_action: Clear next steps for the loan officer
  • Example: Call immediately, send preapproval, schedule follow-up
  • Designed to help convert high-intent leads faster

5. Key Signals

• key_signals: Bullet-point highlights of the most important engagement indicators
  • Examples include:
    • AI readiness scores
    • SMS activity
    • Email engagement
    • Dashboard usage
    • Financial strength (credit score, LTV)

These signals summarize why the lead matters at a glance.

6. Model & Timing

• model: AI model used to generate the score and analysis
• scored_at: Timestamp when the lead was scored
• acknowledged_at: Timestamp when the lead was reviewed (if applicable)

7. Lead (Subscription) Details

Contact Information

• email
• first_name / last_name
• phone

Lead Source & Status

• source: Where the lead originated (e.g., Zapier, website, CRM)
• active: Indicates if the lead is currently active
• created: Timestamp when the lead was created

8. Loan Scenario (Quote Info)

This section describes the borrower’s financial profile and loan scenario:

• credit_score
• loan_type (e.g., Conventional, FHA)
• loan_purpose (Purchase, Refinance)
• loan_amount
• list_price
• debt_to_income
• monthly_income
• loan_term
• property_type
• residency_type

Location Details

• City, State, ZIP, County

These inputs help determine eligibility, pricing, and overall lead quality.

9. Loan Officer Information

• loid: Loan officer ID
• name: Loan officer or company name
• email / phone: Contact details
• nmls: Licensing identifier

This ensures the lead is tied to the correct originator for follow-up.

10. How This Data Is Used

BankingBridge combines:

• Behavioral engagement (emails, SMS, dashboard activity)
• Financial profile (credit, income, LTV)
• AI conversation signals

to produce:

• A real-time intent score
• A clear next action
• A prioritized lead list

11. Key Value for Lenders

• Identify high-intent leads instantly
• Focus on leads most likely to convert
• Reduce missed opportunities
• Improve speed-to-lead and conversion rates

Summary

This dataset gives you a complete picture of each lead by combining who they are, how they behave, and how ready they are to move forward. The result is actionable intelligence that helps loan officers prioritize and close more deals.