Every morning, n8n looks for a real reason to contact each long-term lead, Claude drafts a two-to-four-sentence note in your voice, and nothing goes out until you approve it in Slack
An n8n template by Max (Agentmelt blueprints) runs every day at 6:00 AM and matches long-term leads from your CRM data (the template names Follow Up Boss, kvCORE, or HubSpot) to a reason to write, such as a new listing that fits, a rate change near the lead's budget edge, or a personal date. Guardrails drop anyone contacted recently, unsubscribed, in an active conversation, or in an active CRM stage, with a cap of one touch per 14 days and three per quarter. An Anthropic Claude agent drafts two to four sentences that name the reason, reference what the lead said last time, and ask one question, and it flags low-confidence or fair-housing-sensitive drafts for human review. Flagged drafts go to a review queue, the rest go to a Slack channel for approve or reject, approved messages send through Gmail, and Gmail replies are logged to Google Sheets.
At a glance
- Tools used
- n8nAnthropic ClaudeSlackGmailGoogle Sheets
- Category
- Lead follow-up & inside sales
- Audience
- solo agent, team lead, inside sales agent
- Difficulty
- Builder
- Source
- Nurture real estate leads from Follow Up Boss signals with Claude and Gmail
Max (n8n creator, Agentmelt) · n8n template · Sep 24, 2026 - First step
- Fill in the template's Config node (your name, email, time zone, quiet hours, and the Slack approval channel), leave test mode on, and feed it ten long-term leads with their last conversation notes to see which ones get a draft and which get skipped.
Compliance note: Keep the approval step on, honor STOP and unsubscribe requests on the first reply, and check every draft for fair-housing language before it sends.
Suggested prompt
Written by this directory, inspired by the source. Replace the [bracketed] parts.
You draft a check-in message to a long-term real estate lead. Use only the lead data and the signal below. Write two to four sentences in my voice that (1) name the specific reason I am writing (the listing, the rate change, or the date), (2) reference what the lead told me last time, and (3) ask one question. If the signal is not something this lead would recognize as relevant, return needs_human: true and no draft. Avoid any wording about who belongs in a neighborhood, schools as a proxy for people, or family status. Return JSON with subject, draft, summary, and needs_human. Lead data and signal: [paste]
Get 50 AI Prompts for Real Estate Agents, free
Subscribe to the Daily 5 and download the 48-page PDF right away. The prompts help with:
- Getting more leads from the contacts already in your CRM
- Faster follow-up, so no text or inquiry goes unanswered
- Less time on listing descriptions, paperwork, and showing instructions
- Copy-paste prompts tested by working agents, brokers, and landlords
- Five new real-world AI use cases every weekday morning in the Daily 5
Confirm by email. Unsubscribe in one click. Review every AI draft for accuracy and fair housing compliance before you send it.