name: real-estate-lead-gen description: Automated real estate lead generation pipeline that discovers, qualifies, and delivers seller leads to agents. Use when users want to find motivated sellers (FSBO, expired listings, price reductions, pre-foreclosures), score leads with AI, prospect for agent customers, set up Stripe monetization, or schedule recurring lead delivery via Gmail. Supports any US metro area.

Real Estate Lead Generation Pipeline

Automated pipeline that discovers motivated seller leads, qualifies them with AI scoring, prospects for agent customers, monetizes via Stripe, and delivers results via Gmail on a recurring schedule.

Overview

The pipeline has 6 stages executed in order:

1. Configure — Set target market, ICP, and scoring rubric
2. Monetize — Create Stripe products, pricing tiers, and payment links
3. Discover — Scan listing sites for motivated seller leads
4. Qualify — Score leads in parallel using weighted rubric
5. Prospect — Research and qualify target agent customers
6. Automate — Schedule recurring discovery and delivery

Determine which stages to run based on user request. For first-time setup, run all 6. For recurring execution, run stages 3-4 only.

Stage 1: Configure

Create a working directory and configuration files.

mkdir -p ~/lead_gen_pipeline

Read the ICP template: references/icp_template.md. Customize for the user's target metro area by replacing placeholders ({METRO}, {BOROUGH_LIST}, {NEIGHBORHOOD_LIST}).

Save customized ICP to ~/lead_gen_pipeline/icp.md.

Read the email templates: references/email_templates.md. Customize company name and metro area. Save to ~/lead_gen_pipeline/email_templates.md.

Scoring Rubric (always use these weights):

| Factor | Weight | 10 = Best | 1 = Worst | |--------|--------|-----------|----------| | Motivation | 30% | Must sell (foreclosure, estate, divorce) | Just curious | | Timeline | 25% | Within 30 days | No timeline | | Property Value | 20% | $2M+ | Under $200K | | Engagement | 15% | Responded / open house | No signals | | Competition | 10% | No agent (FSBO) | Exclusive listing |

Final Score = (Motivation × 0.3) + (Timeline × 0.25) + (Value × 0.2) + (Engagement × 0.15) + (Competition × 0.1)

Tiers: HOT (8-10), WARM (5-7.9), NURTURE (1-4.9)

Stage 2: Monetize

Use the Stripe MCP server to create products and pricing. Run these commands in order:

1. create_product: name="Real Estate Lead Intelligence — {METRO}"
2. create_price: Starter $149/mo recurring monthly
3. create_price: Starter $1,490/yr recurring yearly
4. create_price: Professional $299/mo recurring monthly
5. create_price: Professional $2,990/yr recurring yearly
6. create_price: Enterprise $599/mo recurring monthly
7. create_price: Enterprise $5,990/yr recurring yearly
8. create_payment_link: for each monthly price
9. create_coupon: name="FIRSTMONTH50", percent_off=50, duration="once"

Save all IDs and links to ~/lead_gen_pipeline/stripe_config.md.

Tier features:

| Tier | Boroughs | Frequency | Leads/week | |------|----------|-----------|------------| | Starter $149/mo | 1 | Weekly digest | ~20 | | Professional $299/mo | 3 | Daily + weekly | ~50 | | Enterprise $599/mo | All | Real-time + daily + weekly | Unlimited |

Stage 3: Discover

Scan multiple sources for each lead category. Adapt source URLs to the target metro.

Category 1 — Price Reductions: Browse the metro's primary listing site (StreetEasy for NYC, Redfin/Zillow for others) filtered by price drops > 5% in the last 14 days. Collect: address, neighborhood, type, price, drop amount, beds/baths/sqft, brokerage.

Category 2 — FSBO: Browse Zillow FSBO section filtered to the target metro. Collect: address, borough, price, beds/baths/sqft, notes.

Category 3 — Expired/Long DOM: Search for listings with 90+ days on market or recently expired. Collect: address, neighborhood, type, price, DOM, brokerage.

Category 4 — Pre-Foreclosure: Search for lis pendens, NOD filings, or pre-foreclosure listings in the metro. Collect: address, borough, estimated value, filing type.

Save all raw leads to ~/lead_gen_pipeline/daily_leads_raw.md with date header and tables per category.

Target: 20-50 raw leads per scan.

Stage 4: Qualify

Use the map tool to qualify all discovered leads in parallel.

Prompt template for each lead:

You are a real estate lead qualification specialist for the {METRO} market.

Lead to Qualify: {{input}}

Score using this weighted rubric (1-10 each):
- Motivation (30%): 10=must sell, 5=considering, 1=curious
- Timeline (25%): 10=30 days, 7=90 days, 4=6 months, 1=none
- Property Value (20%): 10=$2M+, 8=$1M-2M, 5=$500K-1M, 3=under $500K
- Engagement (15%): 10=responded, 7=open house, 5=active, 3=none
- Competition (10%): 10=no agent (FSBO), 7=expired, 3=has agent

Tasks:
1. Search online for the property to verify details
2. Research comparable recent sales in the neighborhood
3. Assess seller motivation from available signals
4. Calculate weighted final score (1-10)
5. Write a personalized 3-4 sentence outreach script
6. Classify: HOT (8-10), WARM (5-7.9), NURTURE (1-4.9)

Output schema for map tool:

| Field | Type | Description | |-------|------|-------------| | address | string | Full property address | | lead_type | string | Price Reduction / FSBO / Expired / Pre-Foreclosure | | estimated_value | string | Dollar amount | | motivation_score | number | 1-10 | | timeline_score | number | 1-10 | | value_score | number | 1-10 | | engagement_score | number | 1-10 | | competition_score | number | 1-10 | | final_score | number | 1.0-10.0 | | tier | string | HOT / WARM / NURTURE | | outreach_script | string | 3-4 sentence personalized script | | key_signals | string | Comma-separated signals |

Save qualified leads sorted by score to ~/lead_gen_pipeline/daily_leads_qualified.md.

Stage 5: Prospect

Use the map tool to research target agents in parallel.

Identify 10 top-producing agents in the metro by searching for "top real estate agents {METRO} {YEAR}" and collecting names, brokerages, and specialties.

Prompt template for each agent:

Research this real estate agent as a potential customer for an automated
lead generation service that delivers AI-qualified seller leads daily.

Agent: {{input}}

Find: website, email, market focus, estimated annual volume, buying signals
(active marketing, growing team, high volume, awards, thought leadership).

Score purchase likelihood (1-10).
Write a 3-4 sentence outreach email referencing their specific achievements
and offering a free sample report. Service pricing: Starter $149/mo,
Professional $299/mo, Enterprise $599/mo.

Output schema for map tool:

| Field | Type | Description | |-------|------|-------------| | agent_name | string | Full name | | brokerage | string | Company name | | market_focus | string | Borough/neighborhoods | | annual_volume | string | Dollar amount or Unknown | | website | string | URL or Not found | | email | string | Email or Not found | | buying_signals | string | Comma-separated signals | | purchase_likelihood | number | 1-10 | | outreach_draft | string | Personalized email body |

Save to ~/lead_gen_pipeline/agent_prospects.md.

Stage 6: Automate

Schedule recurring tasks using the schedule tool.

Daily lead scan (weekdays 7 AM):

cron: "0 0 7 * * 1-5"
name: "Daily {METRO} Lead Discovery"
prompt: [Run stages 3-4 using files in ~/lead_gen_pipeline/]

The scheduled prompt should instruct Manus to:

1. Read ICP from ~/lead_gen_pipeline/icp.md
2. Read email templates from ~/lead_gen_pipeline/email_templates.md
3. Execute Stage 3 (Discover) for all categories
4. Execute Stage 4 (Qualify) using map tool
5. For HOT leads (8+), send alert via Gmail MCP using Template 1
6. Log activity to ~/lead_gen_pipeline/delivery_log.md

Sending Emails

Use the Gmail MCP server (manus-mcp-cli tool call send_email --server gmail) to deliver leads.

HOT lead alert: Send immediately when a lead scores 8+. Use Template 1 from email templates.

Weekly digest: Compile all qualified leads from the week into a summary table. Use Template 2 from email templates.

Agent outreach: Send personalized prospecting emails to agent prospects. Use Template 3 from email templates.

Always ask user for confirmation before sending emails (confirm_browser_operation).

File Inventory

After full execution, the working directory contains:

| File | Purpose | |------|--------| | icp.md | Ideal Customer Profile and scoring rubric | | email_templates.md | Email templates for alerts, digests, outreach | | stripe_config.md | Stripe product IDs, prices, payment links | | daily_leads_raw.md | Raw discovered leads (updated daily) | | daily_leads_qualified.md | Scored and tiered leads (updated daily) | | agent_prospects.md | Researched agent customers with outreach drafts | | delivery_log.md | Audit log of all deliveries |