🏠 Real Estate & Property Intelligence Pipeline Turn raw RSS feeds into structured Market Intelligence. Scrape, summarize, and analyze Property & Housing news without configuring external API keys. ----- ## What is the Real Estate Intelligence Pipeline? This Actor is a specialized Real Estate Market API and Property News Scraper designed for investors, agents, and prop-tech developers. Unlike standard RSS readers that only give you a headline and a link, this tool uses a sophisticated AI Pipeline to read and understand the market for you. It automatically fetches news from top sources (Realtor.com, HousingWire, The Real Deal), attempts to scrape the full content, and uses a Large Language Model (LLM) to generate structured intelligence: market dynamics, risk sentiment, categorical tagging, and executive summaries. [Image of AI Data Pipeline Architecture] ### Why use this Actor? * 🔋 Batteries Included: You do not need your own OpenAI or Brave Search API keys. We handle the infrastructure; you just pay for the results. * 🧠 AI-Powered Analysis: Every article is processed by an LLM to extract "High Growth" signals, identify "Commercial Distress," and summarize the core facts. * 🛡️ Resilient "Scrape-First" Engine: The actor first attempts to scrape the article directly (free). If blocked, it automatically falls back to a paid Brave Search snippet extraction to ensure you almost always get context for the AI. * 📍 Granular Data: Unlike generic scrapers, this actor extracts specific Locations (e.g., "Austin, TX"), Companies (e.g., "Zillow"), and Hard Statistics (e.g., "Mortgage Rate: 6.1%"). ## What data can this Actor extract? This tool transforms unstructured web content into a structured dataset. Here is the data model you can expect: | Field | Description | | :--- | :--- | | Headline | The original title of the news article. | | Executive Summary | A dense, 2-sentence professional briefing of the event. | | Market Dynamic | The sentiment/impact (e.g., High Growth, Cooling Market, Policy Driven). | | Category | The thematic tag (e.g., Residential Sales, Commercial/Office, PropTech). | | Locations | Specific cities, states, or regions mentioned (e.g., Phoenix, AZ). | | Key Stats | Hard numbers extracted from the text (e.g., Inventory: +5%). | | Companies | Corporate entities involved (e.g., Redfin, BlackRock). | | Extraction Method | Indicates if data came from direct scraping or search fallback. | ## How to use the Real Estate Pipeline This Actor is designed to be "plug and play." 1. Select a Source: Choose a preset (e.g., housingwire, biggerpockets) or select all to aggregate from major industry outlets. You can also provide a custom RSS feed URL. 2. Set Volume: Define the maxArticles you want to analyze (e.g., 10 recent articles) to control costs. 3. Run: Click Start. The Actor will handle the scraping, fallback searching, and AI analysis automatically. > Tip: Enable Run Test Mode in the input settings to generate dummy data. This allows you to test your integration pipelines without incurring any costs. ## Pricing How much does it cost to analyze property news? This Actor uses the Pay-per-event pricing model. This ensures you only pay when the AI successfully generates intelligence. * Usage Cost: You pay a small fee per successful AI Article Analysis (analyze_article_summary event). * Platform Cost: Standard Apify platform usage fees apply for the compute time (duration of the run). Why this model? We pay for the underlying LLM (OpenRouter) and the Search API (Brave) so you don't have to manage subscriptions. The Pay-per-event fee covers these API costs. You are not charged the event fee for articles that fail to be analyzed. ## Input Configuration The input is simple and requires no API keys. * News Source: Select from a dropdown of top Real Estate sites or provide a custom URL. * Max Articles: Limit the number of items to process per run. * Region/Time: (Optional) Filter results if the actor falls back to search (e.g., ensure results are from "US-EN"). * Test Mode: Toggle this to True for zero-cost integration testing. ## Output Example The Actor stores results in the default Apify Dataset. You can export this as JSON, CSV, or Excel. json { "title": "Mortgage Rates Drop Below 6.5%, Sparking Refi Boom", "source_feed": "HousingWire", "published": "2025-10-15T14:30:00Z", "market_dynamic": "High Growth", "category": "Mortgage/Rates", "locations": [ "United States", "California" ], "statistics": [ "Rate: 6.45%", "Applications: +12%" ], "companies": [ "Freddie Mac", "Rocket Mortgage" ], "ai_summary": "The average 30-year fixed mortgage rate fell to 6.45% this week, the lowest level in six months. This drop triggered a 12% surge in refinance applications as homeowners moved to lower their monthly payments.", "method": "scraped", "url": "https://housingwire.com/articles/mortgage-rates-drop..." } ## FAQ & Troubleshooting Is scraping these news sites legal? Our scrapers collect data from publicly available RSS feeds and news articles. We do not extract private user data. However, you should ensure your use of the data complies with copyright laws and the Terms of Service of the target websites. Why did some articles return "Method: search_fallback"? Some property websites have strict anti-scraping protections (Cloudflare, etc.). When our direct scraper is blocked, the Actor automatically queries the Brave Search API to find snippets and summaries of the article from across the web. This ensures you still get an AI analysis even if the direct link was inaccessible. Can I use a custom RSS feed? Yes. Select custom in the Source dropdown and paste any valid RSS XML URL into the customFeedUrl field. I am getting a "Maintenance" warning or empty results? Check if you have runTestMode enabled. If you are running this locally or via API, ensure you are respecting the schema. If the issue persists, please open an Issue in the Console tab. ## Advanced: The Intelligence Pipeline For developers integrating this into a larger system (e.g., a programmatic SEO blog or investment dashboard), it helps to understand the flow: 1. Ingest: Fetch N items from the RSS feed. 2. Deduplicate: Check against previous runs to avoid analyzing the same story twice (saving you money). 3. Scrape (Tier 1): Attempt fast, direct HTML extraction using standard headers. 4. Fallback (Tier 2): If Tier 1 fails (403/429), perform a semantic search for the headline using Brave Search. 5. Synthesize: Feed the best available text context to the LLM for structured JSON extraction.