Real-Estate Buyer Matching and Notification

1. The Problem

Real estate agencies maintain two parallel databases that are almost never systematically matched: property listings and registered buyer criteria. When a new property comes to market, the agent who happens to be working that day may remember two or three clients whose criteria it fits — and misses twelve others whose registered requirements are an exact match. Every unmade match is a lost sale. Every late notification is a competing agency getting there first. The matching problem has been solved manually for decades. It should not still be manual.

2. Integration & Webhook Setup

Follow the exact steps below to configure and deploy this automation inside your OpenClaw workspace.

1. 1Create a new agent named `real-estate-matching`. Connect your CRM (most estate agencies use Reapit, AgentOS, or Salesforce — all supported) and your property portal feed (Rightmove and Zoopla both offer data feeds for connected agencies).
2. 2Configure your `MATCHING_CRITERIA` taxonomy: define the buyer preference fields used in your CRM (minimum bedrooms, maximum price, preferred postcodes, required features like garage or garden) and map them to the property attribute fields in your listing database. The agent matches on this structured data, not free-text descriptions.
3. 3Set up your `COMPARABLE_SALES_SOURCE`: the agent pulls comparable transaction data from Land Registry (free, 3-month lag) and/or your integrated portal's sold price data (real-time, requires portal data feed agreement). Configure the comparison radius and the attributes used for comp selection.
4. 4Build your notification templates: buyer alerts should feel like a personal recommendation from the agent, not an automated portal alert. Include the agent's name, a specific line referencing why this property matches the buyer's registered criteria, and a direct booking link.
5. 5Configure your `PRICE_POSITION_MONITOR`: set the enquiry threshold (default: fewer than 3 enquiries in 7 days triggers a price alert) and the market data package the agent assembles for the vendor conversation — price per square foot vs comparable actives, days on market distribution, and recent sold prices within 0.5 miles.

3. The Context Payload (context.json)

Save this file as: .openclaw/agents/real-estate-buyer-matching-and-notification/context.json

{
  "automation_id": "25",
  "title": "Real-Estate Buyer Matching and Notification",
  "level": 3,
  "tier": "Enterprise Playbook",
  "setup_time": "1 week",
  "estimated_api_cost": "~$50–$100/mo",
  "client_price_range": "$2,000–$10,000",
  "agents": [
    {
      "role": "orchestrator",
      "model": "claude-3-5-sonnet",
      "temperature": 0.2,
      "max_tokens": 4096
    }
  ],
  "memory": "session",
  "output_format": "structured_json",
  "human_review_gate": true,
  "documentation_standard": "required"
}

4. Execution Commands

Run these commands from your openclaw-workshop/ directory to validate, test, and schedule this automation. Commands are taken directly from The OpenClaw Income Engine, Appendix G.

# ── STEP 1: Map CRM fields to property fields ──
$ openclaw run real-estate-matching --map-fields --dry-run
# Fix any mapping mismatches in context.json before proceeding.
$ 
# ── STEP 2: Test with a single listing against all registered buyers ──
$ openclaw run real-estate-matching --listing-id <crm_property_id> --dry-run
$ 
# ── STEP 3: Review match output — check score accuracy ──
# Are the right buyers being matched? Adjust weights in context.json if needed.
$ 
# ── STEP 4: Activate portal feed webhook ──
$ openclaw webhook activate real-estate-matching --event new_listing_or_status_change
$ 
# ── STEP 5: Run price position monitor for all listings on market >7 days ──
$ openclaw run real-estate-matching --mode price-monitor --days-on-market 7
$ 
# ── Run matching for a specific listing on demand ──
$ openclaw run real-estate-matching --listing-id <id> --mode matching
$ 
# ── Generate a comparable sales report for a specific property ──
$ openclaw run real-estate-matching --listing-id <id> --mode comparables
$ 
# ── View all buyers who have not been matched to any listing in 30 days ──
$ openclaw run real-estate-matching --unmatched-buyers --days 30