Lilly Tech SystemsAI Forecasting & Scoring Intermediate
HubSpot's AI forecasting and scoring features use machine learning to predict which leads are most likely to convert, which deals will close, and what revenue you can expect. These predictions help marketing and sales teams prioritize their efforts and plan more accurately.
Predictive Lead Scoring
HubSpot's predictive lead scoring analyzes hundreds of data points from your CRM to automatically score contacts by their likelihood to convert. Unlike manual lead scoring, the AI model learns from your actual conversion data.
| Signal Category | Data Points Analyzed |
|---|---|
| Engagement | Email opens, clicks, page visits, form submissions, content downloads |
| Firmographic | Company size, industry, revenue, location, technology stack |
| Behavioral | Website visit frequency, pages viewed, time on site, return visits |
| Demographic | Job title, seniority, department, education |
Deal Forecasting
AI deal forecasting provides more accurate pipeline predictions than simple stage-based estimates by analyzing deal velocity, engagement signals, and historical close rates per segment.
Revenue Forecasting
- AI-powered forecast: ML models analyze pipeline data, historical close rates, and seasonal patterns to predict revenue
- Forecast categories: Deals are classified as commit, best case, or pipeline with AI-assigned probabilities
- Trend analysis: AI identifies whether your pipeline is trending above or below target and flags risks
- Scenario modeling: Explore how changes in close rates, deal sizes, or pipeline volume would affect revenue projections
Data requirements: Predictive scoring works best with at least 500 contacts and 100 conversions in your CRM. The more historical data available, the more accurate the AI predictions become.
Action step: Enable predictive lead scoring in HubSpot Settings under Properties. Review the top-scoring leads and compare with your sales team's assessment. Use the alignment (or misalignment) to calibrate how you use the scores.