AI-Powered Lead Routing with Make.com, OpenAI, and HubSpot

Introduction

Lead routing in HubSpot often becomes a bottleneck. New inquiries arrive, get triaged by a human, and sometimes fall through the cracks due to speed or context gaps. This advanced setup uses Make.com to orchestrate a real-time AI driven routing flow. OpenAI analyzes the lead context and suggests an owner, then Make.com updates HubSpot and notifies the right person in Slack. The result is faster handoffs, consistent ownership, and a complete audit trail.

By the end of this guide you will know how to wire Make.com, OpenAI, and HubSpot to automatically classify, assign, and notify on new leads.

What You'll Need

• Make.com account with access to scenarios (formerly Integromat)
• HubSpot account with API access and the ability to assign owners
• OpenAI API access (GPT-4 or higher) with a valid key
• A Slack workspace with a channel for sales leads (e.g., #sales-leads)
• Optional: Google Sheets or Airtable for lead routing logs

How It Works (The Logic)

Trigger: a new contact is created in HubSpot or lifecycle stage becomes NEW. Make.com pulls data, passes data to OpenAI to generate a lead score and owner recommendation. Then Make.com updates the HubSpot contact with the score and owner, adds a contextual note, and posts a summary to Slack. Finally, it can log the run in a sheet for auditing.

Step-by-Step Setup

1. Create a new scenario in Make.com and add the HubSpot module set to Watch Contacts, with a filter for lifecycle stage equals NEW or date created within the last 5 minutes. Map fields you care about such as email, firstname, lastname, company, job title, website, and recent inquiry text.

2. Add an OpenAI module in the scenario. Configure a prompt that feeds the contact data into an AI task. Example prompt:

   • Objective: “Given the contact data, produce a JSON object with fields: lead_score (0-100) and owner_email (HubSpot owner email).”
   • Inputs: company, job_title, industry, company_size, inquiry_text, and source.
   • Model: gpt-4 or higher. Output: a JSON string with lead_score and owner_email.

3. Add a HubSpot module to fetch or verify the contact by email. Use a Search or Get Contact by Email action to ensure you update the correct CRM record.

4. Add a HubSpot module to Update a Contact. Map the fields from the AI output: set lead_score property to the numeric value and owner_id or owner_email to the HubSpot owner field. Include any additional fields such as lifecycle stage and last_contacted timestamp.

5. Add a HubSpot module to Add a Note. Include the AI advised rationale and the final routing decision. This creates an audit trail right in the contact timeline.

6. Add a Slack module. Post to your sales channel with a concise summary:

   • Text: New lead assigned to {owner_name} with score {lead_score}. Source: {source}. Name: {firstname} {lastname}. Company: {company}.
   • Include a quick link to the HubSpot contact for follow up.

7. Optional: Add a Google Sheets or Airtable module to log the run. Create a row with fields like timestamp, lead_email, lead_score, owner_email, and outcome.

8. Add a final step to trigger a quick test. Create a test contact in HubSpot or use a sample payload. Run the scenario and verify the Slack message and HubSpot updates reflect the AI result.

9. Turn on scheduling. Set the scenario to run every 5 minutes or at a cadence that fits your lead volume. Validate error handling and retries so transient API issues don’t stall routing.

Field mappings you’ll likely use

• HubSpot Watch: email -> email, firstname -> firstname, lastname -> lastname, company -> company, job_title -> job_title, inquiry_text -> notes
• OpenAI Output: lead_score -> numeric, owner_email -> string
• HubSpot Update Contact: lead_score -> custom property on contact, owner -> owner_id
• Slack Message: use variables {firstname}, {lastname}, {lead_score}, {owner_name}, {source}
• Google Sheets: timestamp, lead_email, lead_score, owner_email, outcome

Real-World Business Scenario

A mid market SaaS company used this setup to automate triage for 300 new leads per week. The AI routing reduced response time from hours to minutes and raised the percentage of leads assigned within 5 minutes from 42% to 88%. Sales reps started their follow ups faster, and the audit notes gave the operations team clear visibility into how each lead was routed and why.

If you want to see how we implement similar end-to-end automations for real teams, check our work with Typeform to HubSpot and Monday.com integrations in a recent client project referenced in our case studies.

Common Variations

• Add a PDF proposal trigger: after lead routing, generate a tailored proposal using a template in PandaDoc or DocuSign and attach it to the HubSpot deal.
• Refine routing rules by territory or product interest and include a separate Slack channel or email distribution per product line.
• Add a feedback loop: if owners mark a lead as unqualified, send a notification to the SDR team and adjust lead score criteria in OpenAI prompts.

Scale Your Lead Routing

This setup gives your team a tangible speed and consistency advantage. You can customize prompts, scoring, and owner assignments as you learn what data truly predicts deal value in your market. Olmec Dynamics loves building these AI enhanced workflows for fast growing teams. If you want help tailoring this to your business, we can build it for you — learn more about our automation services at Olmec Dynamics.