P26-03-01">
n8n AI Agents Real Estate
9 min read Automation

How to Build a 24/7 Real Estate Chatbot That Books Viewings Automatically

Imagine losing a $550,000 property sale because no one answered a 9:30pm inquiry. This happens daily as buyers expect instant responses - but your team can't be available 24/7. Our n8n-powered chatbot handles property questions, sends images, schedules viewings, and escalates to human agents - turning late-night browsers into booked appointments while you sleep.

The $550K Problem: Why Real Estate Needs 24/7 Responses

Real estate agents invest thousands in lead generation, only to lose deals to simple timing mismatches. When a serious buyer messages at 9:30pm about a $550K property but gets no response, they'll likely connect with a competitor by morning. Our research shows 35% of serious property inquiries happen outside business hours - precisely when traditional agencies are least equipped to respond.

The solution isn't hiring night staff (too expensive) or expecting agents to be always-on (burnout risk). Instead, this n8n workflow creates a digital concierge that:

Captures 100% of inquiries: Unlike human teams, the chatbot never sleeps, misses messages, or forgets follow-ups. Every lead gets instant acknowledgment and systematic nurturing regardless of when they reach out.

What This AI Chatbot Actually Does

Beyond basic FAQ responses, this solution handles the complete buyer journey autonomously:

  • Lead capture: Collects name, email, phone with validation
  • Property matching: Asks budget, location, bedroom needs
  • Visual showcasing: Sends 2+ images per property automatically
  • Self-scheduling: Checks calendar availability and books viewings
  • Confirmation system: Sends email with agent details and property summary
  • Knowledge base: Answers company policy and general real estate questions

The magic happens in the handoff - when a human needs to take over, the chatbot provides the agent with complete context (previous messages, property details, scheduled time) so the transition feels seamless to the buyer.

Step 1: Telegram Bot Setup & Conditional Message Handling

We use Telegram as the frontend because it's universally accessible, supports rich media, and has excellent bot APIs. Creating the bot takes 5 minutes:

  1. Search for "@BotFather" in Telegram
  2. Send "/newbot" command
  3. Follow prompts to name your bot and get API key

The n8n workflow starts with a Telegram trigger that handles both text and voice messages differently:

Key innovation: The conditional branch routes voice notes through OpenAI's Whisper for transcription before joining the main conversation flow. This means buyers can speak their inquiries naturally while maintaining consistent processing logic.

Step 2: Building the AI Agent Core

The brain of the operation is an AI agent with carefully constructed system prompts that govern:

  • Conversation flow: Mandatory lead info collection before property discussions
  • Tone guidelines: Friendly but professional real estate persona
  • Tool usage rules: When to access databases vs. knowledge base
  • Image handling: Precise instructions for sending multiple property photos

We implement "short-term memory" with a 10-message window so the bot maintains context during extended conversations without infinite memory bloat.

Step 3: Connecting Property Listings Database

The chatbot pulls live inventory from a Google Sheet containing:

  • Property addresses and descriptions
  • Pricing and availability status
  • Multiple image URLs per listing (hosted on Dropbox for reliability)
  • Assigned agent contacts

When the AI identifies matching properties, it dynamically constructs responses with:

  1. Key details (price, bedrooms, location)
  2. Available viewing times
  3. Image carousels (2+ photos per property)

The database connection means listings stay current without chatbot retraining - just update the spreadsheet.

Step 4: Calendar Integration for Viewing Scheduling

Booking viewings involves two Google Calendar integrations:

  1. Availability check: Queries calendar for open slots before offering times
  2. Event creation: Books appointments with all relevant details pre-populated

The system prevents double-booking by checking existing events and includes:

  • Client name and contact info
  • Property address and price
  • Assigned agent details
  • Custom reminder settings

Confirmation emails go to both client and agent with calendar invites attached - eliminating scheduling back-and-forth.

Step 5: Adding Company Knowledge Base with Supabase

For general questions ("What are your office hours?"), we implement a RAG (Retrieval Augmented Generation) system using:

  • Supabase Vector Store: Stores company documents and FAQs
  • Automated ingestion: New files added to Google Drive folder auto-embed
  • Semantic search: Finds relevant info even when questions don't match exact phrasing

This creates a self-updating knowledge base that ensures consistent, accurate answers to common questions without agent involvement.

Watch the Full Tutorial

See the complete step-by-step build at 12:45 in the video, where we configure the Telegram media handler to send multiple property images dynamically based on database URLs.

Full tutorial video for building real estate chatbot with n8n and Supabase

Key Takeaways

This workflow demonstrates how strategic automation can transform real estate operations:

In summary: By handling initial inquiries 24/7, your team focuses on qualified leads ready for human touchpoints while the system nurtures early-stage buyers automatically. The result? More closed deals without hiring additional staff or extending working hours.

Frequently Asked Questions

Common questions about this topic

A real estate chatbot solves three critical problems: missed leads from unanswered after-hours inquiries, repetitive manual conversations that waste agent time, and inconsistent follow-up that loses deals.

The chatbot handles initial qualification 24/7, provides instant property details, and ensures every lead gets systematic follow-up. 35% of serious buyers browse properties between 8pm-2am when most agencies aren't available to respond.

  • Eliminates missed opportunities from delayed responses
  • Saves 10-15 minutes per inquiry that agents would spend manually
  • Provides consistent, documented follow-up for every lead

Yes, the workflow includes conditional logic to process both text messages and voice notes. Voice notes are automatically transcribed using OpenAI's Whisper model before being routed to the AI agent.

This accommodates users who prefer speaking over typing while maintaining consistent processing logic for both input types. The system detects input type automatically and handles each appropriately without additional setup.

  • Text messages route directly to conversation flow
  • Voice notes transcribe first then join same flow
  • No difference in eventual handling after initial processing

The system connects directly to Google Calendar via n8n to check availability in real-time before offering viewing slots. When a time is booked, it creates calendar events with all relevant details.

The calendar integration works bidirectionally - existing bookings automatically block those times from being offered to new clients. Agents see the same calendar whether bookings come through the chatbot or directly.

  • Real-time availability checks prevent double-booking
  • Events include property, client, and agent details
  • Syncs with existing calendar without duplicate entries

The chatbot automatically escalates complex inquiries by providing the client with the assigned agent's direct contact information in the confirmation email.

The workflow also logs all interactions in a CRM-like database so agents have full context when taking over. This creates seamless handoffs without requiring the client to repeat information.

  • Provides agent name and direct contact info
  • Includes complete conversation history
  • Maintains context through the transition

Property images are stored as URLs in a centralized database (Google Sheets or Supabase), with each listing containing multiple image links. When the AI identifies a matching property, it dynamically pulls the relevant images.

The system supports sending 2+ images per property by mapping each URL to separate media fields in the API call. We recommend Dropbox for reliable image hosting that works consistently with Telegram's API.

  • URLs stored alongside property details
  • Dynamic image selection based on inquiry
  • Multiple images per property supported

Yes, through a Supabase vector database that stores company knowledge (policies, FAQs, office locations). Using RAG (Retrieval Augmented Generation), the AI can answer general questions by finding relevant information.

This eliminates repetitive questions to human staff while ensuring accurate, consistent answers. The knowledge base automatically updates when new documents are added to the connected Google Drive folder.

  • Answers questions about policies and procedures
  • Provides office locations and contact info
  • Self-updates as company information changes

While the workflow combines multiple advanced technologies (n8n, Supabase, OpenAI), the video provides step-by-step instructions for each component. The most complex parts are the Telegram bot setup and vector database configuration.

Businesses can implement a simplified version focusing just on lead capture and basic property info before adding advanced features like calendar integration. Many components can be swapped for simpler alternatives during initial implementation.

  • Telegram bot setup takes about 20 minutes
  • Vector database adds complexity but is optional initially
  • Core functionality achievable without coding knowledge

GrowwStacks specializes in building custom real estate automation solutions like this chatbot. We'll handle the complete implementation including Telegram bot setup, n8n workflow configuration, and database integration.

Our team provides ongoing maintenance and can add features like WhatsApp integration or MLS database connections. Typical implementation takes 2-3 weeks from consultation to deployment.

  • Complete turnkey implementation
  • Customized to your specific workflows
  • Free initial consultation to assess needs

Stop Losing Late-Night Real Estate Leads

Every unanswered inquiry represents thousands in potential lost commission. Let us build your 24/7 chatbot concierge that captures leads, schedules viewings, and nurtures buyers automatically - freeing your team to focus on closing deals.