What this automation does
When a new lead enters your CRM — from a form submission, demo request, or inbound email — this automation enriches the lead with company data, analyzes their behavior signals, and assigns a score from 0-100. High-scoring leads get routed to sales immediately with a summary of why they are qualified.
This replaces the manual process of sales reps scrolling through lists of unqualified leads. Instead, your team focuses only on prospects that match your ideal customer profile and show buying intent. Most teams see a 2-3x improvement in conversion rates within the first month.
Tools you need
- HubSpot or Salesforce CRM: Where leads are stored and scored
- OpenAI API: Analyzes lead data and assigns scores ($0.02-0.05 per lead)
- Make or Zapier: Connects your CRM to OpenAI and updates lead records automatically
How to set it up
Step 1: Define your ideal customer profile. List the attributes that make a lead qualified — company size, industry, job title, budget range, and specific behaviors like visiting your pricing page or downloading a whitepaper. Assign weights to each factor.
Step 2: Create a Make scenario triggered when a new contact is created in your CRM. Pull the lead's name, email, company, and any form responses. Use an enrichment API like Clearbit or Apollo to add company size, industry, and revenue data.
Step 3: Send the enriched data to OpenAI with a prompt that includes your ICP criteria and scoring rubric. Ask it to return a score from 0-100, a qualification status (hot, warm, cold), and a one-sentence reason explaining the score.
Step 4: Update the CRM record with the AI score and qualification status. For leads scoring 70 or above, send a Slack notification to the assigned sales rep with the lead summary. Test with 50 recent leads and compare AI scores against actual outcomes to calibrate.
Cost breakdown
| Item | Cost | Notes |
|---|---|---|
| Make or Zapier | $20-$40/mo | Based on lead volume |
| OpenAI API | $15-$30/mo | ~$0.03 per lead at 500-1000 leads/mo |
| Enrichment API | $0-$50/mo | Clearbit, Apollo, or similar — optional |
| Setup time | 45-90 min | One-time |
| Total monthly | $35-$120/mo | Saves 5+ hours per week of manual qualification |
Frequently asked questions
After calibration, AI lead scoring typically matches or exceeds human accuracy. The key advantage is consistency — AI applies the same criteria to every lead without fatigue or bias. Most teams reach 85%+ accuracy within two weeks of tuning the scoring prompt.
Yes. You can run AI scoring alongside your existing point-based CRM scoring. Many teams use AI scores as an additional data point rather than replacing their current system entirely, then gradually shift as confidence builds.