🎣 Label genuine sales leads in Gmail with OpenAI GPT-4.1-mini

⚡ 0 views · 🎣 Lead Generation & Enrichment

Description

Quick overview

Small business owners, freelancers and solo operators whose Gmail mixes real inbound sales interest with newsletters, vendor pitches and other noise. This workflow uses OpenAI to spot genuine sales leads and labels them in Gmail. It only adds a label; it never sends, deletes or drafts email.

How it works

1. Gmail Trigger checks for new emails every minute.
2. Settings holds your business name, OpenAI model, the Gmail label to apply (default AI/Lead) and optional notes.
3. OpenAI (gpt-4.1-mini by default) decides whether each email is a genuine sales lead (someone who wants to buy from you) and returns strict JSON with is_lead, a short summary and a reason.
4. When is_lead is true, the workflow applies your label. If the label does not exist yet, it is created once (with its parent, AI) and then applied.
5. Cold pitches, newsletters and invalid AI output are skipped (treated as not-a-lead).
6. It only adds a label. It never sends, deletes or drafts email.

Setup

1. Attach your Gmail OAuth2 credential to Gmail Trigger, Get Gmail Labels, Create Lead Label, Re-fetch Gmail Labels and Apply Gmail Label, and your OpenAI credential to Classify Email.
2. Open Settings: set business_name, keep or change openai_model (default gpt-4.1-mini), optionally rename lead_label (the label is created automatically; for a nested label, write its full path, for example AI/Lead), and add plain-English classification_notes if you need them (for example, “Inbound demo requests from the website are always leads”).
3. Send yourself a test enquiry, click Execute workflow, confirm the AI/Lead label appears, then flip the workflow to Active.

Requirements

• n8n (Cloud or self-hosted, version 1.x or 2.x; built on 2.43.5)
• A Gmail / Google Workspace account
• An OpenAI API key (usage billed by OpenAI; with gpt-4.1-mini usually a fraction of a cent per email)

Customization

• Change what counts as a lead in the Classify Email system prompt.
• Rename the label via lead_label in Settings. Use the full path for nested labels (for example AI/Lead or Sales/Hot). Missing labels are created automatically.
• Raise or lower max_body_chars to control how much of each email is sent to the model.
• Swap the OpenAI node for another provider’s chat model with an LLM Chain if you prefer.

🔗 Nodes Used

Gmail, Gmail Trigger, OpenAI

📥 Import

Download workflow.json and import into n8n: Workflow menu → Import from File

📖 Importing guide · 🔑 Credential setup