Lead Scoring Model Builder
Marketing AutomationadvancedClaude 3.5 Sonnet or GPT-4o. Claude excels at building systematic frameworks and explaining complex scoring logic clearly. GPT-4o offers faster processing for large datasets. Both handle the multi-variable weighting and business logic required. Use Claude if you need detailed reasoning; use GPT-4o for speed when refining existing models.
When to Use This Prompt
Use this prompt when you need to build or rebuild a lead scoring system that aligns sales and marketing. It's essential when you have inconsistent lead quality, sales complaints about lead relevance, or when you're scaling and need systematic prioritization. Perfect for teams transitioning from gut-feel qualification to data-driven scoring.
The Prompt
You are a lead scoring strategist helping me build a predictive lead scoring model for [INDUSTRY] companies. I need to create a quantitative framework that identifies which leads are most likely to convert to customers.
## Business Context
- Target customer profile: [DESCRIBE TARGET CUSTOMER]
- Average deal size: [DEAL VALUE]
- Sales cycle length: [CYCLE LENGTH]
- Current conversion rate: [CONVERSION %]
- Primary revenue model: [BUSINESS MODEL]
## Available Data Points
Provide scoring weights for these lead attributes:
- Demographic: [LIST AVAILABLE DEMOGRAPHIC DATA]
- Firmographic: [LIST COMPANY ATTRIBUTES]
- Behavioral: [LIST TRACKED BEHAVIORS]
- Engagement: [LIST ENGAGEMENT METRICS]
- Intent signals: [LIST INTENT DATA SOURCES]
## Scoring Requirements
1. Create a 0-100 point scale with clear thresholds
2. Define minimum score for sales handoff: [TARGET THRESHOLD]
3. Identify 3-5 disqualifying factors that override high scores
4. Build decay rules for stale leads (no engagement in [X] days)
5. Account for seasonal variations in [INDUSTRY]
## Output Format
Provide:
1. Scoring matrix with point allocations for each attribute
2. Calculation methodology (linear, weighted, or tiered)
3. Sales-ready lead grade definitions (A/B/C/D)
4. Implementation checklist for [CRM PLATFORM]
5. Monthly review metrics to validate model accuracy
6. Risk factors that indicate low-quality leads despite high scores
## Constraints
- Prioritize data we already collect (avoid new data dependencies)
- Ensure sales team can understand and trust the scoring logic
- Build in flexibility to adjust weights based on quarterly performance
- Account for [SPECIFIC MARKET CONDITION OR CHALLENGE]
Create a production-ready lead scoring model I can implement immediately.
Get the Full AI Marketing Learning Path
Courses, workshops, frameworks, daily intelligence, and 6 proprietary tools — built for marketing leaders adopting AI.
Trusted by 10,000+ Directors and CMOs.
Tips for Best Results
- 1.Provide actual historical conversion data if available—the model improves dramatically with real performance benchmarks rather than assumptions.
- 2.Include 2-3 specific disqualifying factors unique to your business; generic models fail because they don't account for your deal-breakers.
- 3.Request a 'validation plan' in the output—ask the AI to suggest which metrics to track monthly to prove the model works before full rollout.
- 4.Test the model on 50-100 recent leads before implementation; ask the AI to predict their outcomes and compare against actual results.
Example Output
# Lead Scoring Model for B2B SaaS
## Scoring Scale: 0-100 Points
**Sales Handoff Threshold: 50+ points**
## Demographic Scoring (25 points max)
- Job title match (0-10): VP/C-level = 10, Manager = 7, Individual Contributor = 3
- Department alignment (0-10): Primary buyer = 10, Influencer = 6, Unrelated = 0
- Company size (0-5): Target range = 5, Adjacent = 2, Outside = 0
## Firmographic Scoring (25 points max)
- Industry vertical (0-10): Tier 1 target = 10, Tier 2 = 6, Other = 2
- Annual revenue (0-10): $50M-$500M = 10, $10M-$50M = 7, Other = 3
- Growth rate (0-5): >20% YoY = 5, 10-20% = 3, <10% = 0
## Behavioral Scoring (30 points max)
- Website visits (30 days): 5+ visits = 8, 2-4 = 5, 1 = 2
- Content engagement: Downloaded 3+ assets = 10, 1-2 = 6, None = 0
- Email engagement: Open rate >40% = 8, 20-40% = 5, <20% = 2
- Demo request: Requested = 12, Attended = 15
- Pricing page visits: 3+ visits = 7, 1-2 = 4
## Intent Signals (20 points max)
- Third-party intent data: High intent = 10, Medium = 6, Low = 2
- Keyword search activity: Competitor research = 8, Solution research = 5
- LinkedIn profile updates: Recent changes = 4
## Lead Grades
- **Grade A (80-100)**: Immediate sales outreach, high conversion probability
- **Grade B (60-79)**: Sales development sequence, nurture in parallel
- **Grade C (50-59)**: Marketing nurture, monitor for grade improvement
- **Grade D (<50)**: Continue nurturing, re-score monthly
## Disqualifying Factors (Override scoring)
- Competitor employee
- Unsubscribed from communications
- No engagement in 90+ days (automatic decay to <40 points)
- Explicitly stated "not interested"
- Outside target geography
## Implementation Timeline
- Week 1: Configure scoring rules in CRM
- Week 2: Backtest against historical conversion data
- Week 3: Train sales team on lead grades
- Week 4: Launch with monitoring dashboard
Related Prompts
Related Reading
Get the Full AI Marketing Learning Path
Courses, workshops, frameworks, daily intelligence, and 6 proprietary tools — built for marketing leaders adopting AI.
Trusted by 10,000+ Directors and CMOs.