Redfin Listing and Estimate Scraper
Collecting public Redfin pages by hand does not scale: every listing is a separate page, prices move, and the site is built to resist automated traffic.
A hosted scraper API replaces the browser fleet and proxy pool with one HTTP call. You pass a Redfin URL, the service renders the page, rotates residential proxies, and returns HTML, JSON or Markdown that you can parse or feed to a model. This page explains what comes back, how to choose an extraction mode, and how request options affect credit use.
What you can collect from Redfin
A Redfin listing page is a dense record. The scraper is built to return the address, list price, sold price, Redfin Estimate, bedroom and bathroom counts, square footage, days on market, agent details and the image gallery. Sold-price data is available across US zip codes, which is what makes Redfin useful for market-comp research rather than just single-property lookups.
Three collection patterns cover most teams:
- Pricing research. Pull list and sold prices for a set of neighborhoods to build or validate a pricing model.
- Amenity and feature data. Capture what a property offers so listings can be compared or ranked.
- Closed-sale tracking. Follow sold prices over time to see how a local market is moving.
Each pattern uses the same request shape; only the target URLs and the extraction rules change.
How a request works
The workflow has four steps. First, create an account: you receive an API key and a starting credit allowance, and the exact size of that allowance is set by the current plan terms. Second, choose a client. SDKs are published for Python, Node.js, Go, PHP, Ruby and Java, and plain cURL works too, so use whatever your stack already uses. Third, send a GET request with your API key and a Redfin URL on redfin.com; JavaScript is rendered, residential proxies rotate, and rendered HTML comes back. Fourth, add options only when needed: render_js for JavaScript-heavy pages, stealth_proxy for CAPTCHA-heavy pages, country_code for geo-targeting, and ai_extract_rules for natural-language extraction. The dashboard includes a request builder, so you can test a call there and then copy the generated snippet into your codebase, which avoids debugging authentication and parameters at the same time.
Choosing an output mode
The API can return more than one shape, and the right choice depends on what happens after the request.
- Rendered HTML is the default. Use it when you already have a parser or want full control over which elements you read.
- Structured JSON via `extract_rules` lets you name CSS or XPath selectors and get back only the fields you asked for. Best when the schema is stable and you want a small payload.
- Structured JSON via `ai_extract_rules` describes the fields in natural language and lets the service locate them. Best when the page layout shifts or you are prototyping a schema.
- Markdown turns a listing page into clean text suitable for an LLM prompt or a retrieval pipeline.
- Screenshots capture how a real browser renders the page, useful for visual checks or evidence.
A practical rule: start with ai_extract_rules to discover which fields are reliably present, then move to extract_rules once the schema is settled. That keeps early iteration fast without locking in brittle selectors.
Field and control map
It helps to separate what you get back from what you send. The table below groups the two sides so a request can be debugged one layer at a time.
| Layer | Item | What it controls |
|---|---|---|
| Returned data | Address, list price, sold price, Redfin Estimate | Core pricing fields on a listing |
| Returned data | Beds, baths, square footage, days on market | Property attributes |
| Returned data | Agent details, image gallery | Contact and media |
| Request option | render_js | Whether JavaScript is executed before capture |
| Request option | stealth_proxy | Whether stealth proxies are used for harder pages |
| Request option | country_code | Which country or territory the request appears to come from |
| Request option | extract_rules | CSS or XPath selectors for structured output |
| Request option | ai_extract_rules | Natural-language field descriptions for AI extraction |
If a field is missing, check the returned HTML first: the value may not be on the public page at all. If the page is empty or blocked, the problem is upstream of parsing.
Worked example: one listing into a comps row
Suppose you want a single row for a recently sold property. You send the listing URL with render_js enabled and extract_rules naming the fields you need. The response gives you address, list price, sold price, Redfin Estimate, beds, baths, square footage and days on market. You write those into one row of a comps table.
Repeat for each comparable property in the same zip code. Because sold-price data is published across US zip codes, the same request shape works whether you are sampling one neighborhood or many. The row is only as good as the page: if a listing has no sold price yet, that cell is empty rather than zero, and your downstream logic should treat it that way.
Telling failures apart
Three failure modes look similar in a log but need different fixes.
- Rendering failure. The HTML comes back but the listing values are absent because scripts did not run. Fix: enable
render_js. - Blocking failure. The response is a challenge page or an empty shell. Fix: enable
stealth_proxyand, if needed, adjustcountry_code. - Parsing failure. The HTML contains the values but your selectors miss them. Fix: inspect the rendered HTML and update
extract_rules, or switch toai_extract_ruleswhile you re-derive the selectors.
Checking the raw HTML before changing request options prevents the common mistake of adding proxies to a page that was already returning data.
Credits and plans
New accounts start with a free credit allowance, and the current amount is stated on the pricing page. Paid tiers scale up for more credits and concurrency. Credit cost per request depends on the options you enable: JavaScript rendering costs 5 credits, premium proxies cost 10 or 25 depending on whether JS rendering is on, and stealth_proxy costs 75. API credits are valid for one month and do not roll over. The plan comparison below is a dated reference; prices and inclusions can change, so check the current pricing page before committing to a tier. All prices shown are exclusive of VAT. If you are unsure which tier fits, the free credits are enough to measure real credit consumption on your own URLs before choosing a plan.
Compliance and responsible use
Public-facing pages can usually be collected, but the target site's terms of service should be reviewed before deployment, and post-login content should be avoided. Whether a particular use is permitted depends on those terms, on applicable law in your jurisdiction, and on the nature of the data, so treat legal review as a step in your own process rather than something this page can settle. Two practical limits matter for Redfin work: the site's own anti-bot measures can change without notice, and listing data itself changes as properties are sold or relisted. A pipeline that re-checks a sample of URLs on a schedule will catch both kinds of drift earlier than one that trusts a single historical pull.
| Plan | Price | API credits | Concurrent requests | JS rendering | Rotating & premium proxies | Geotargeting | Screenshots & extraction rules | Priority email support | Dedicated account manager | Team management |
|---|---|---|---|---|---|---|---|---|---|---|
| Hobby | $19 /mo | 75,000 | 25 | Included | Included | Included | Included | Not included | Not included | Not included |
| Freelance | $49 /mo | 250,000 | 50 | Included | Included | Included | Included | Not included | Not included | Not included |
| Startup | $99 /mo | 1,000,000 | 100 | Included | Included | Included | Included | Not included | Not included | Not included |
| Business | $249 /mo | 3,000,000 | 200 | Included | Included | Included | Included | Included | Included | Included |
| Business + | $599 /mo | 8,000,000 | 400 | Included | Included | Included | Included | Included | Included | Included |
| Enterprise 14 | $999 /mo | 14,000,000 | 500 | Included | Included | Included | Included | Included | Included | Included |
| Enterprise 24 | $1,599 /mo | 24,000,000 | 600 | Included | Included | Included | Included | Included | Included | Included |
| Enterprise 41 | $2,399 /mo | 41,000,000 | 700 | Included | Included | Included | Included | Included | Included | Included |
| Enterprise 69 | $3,599 /mo | 69,000,000 | 800 | Included | Included | Included | Included | Included | Included | Included |
| Enterprise 120 | $5,799 /mo | 120,000,000 | 900 | Included | Included | Included | Included | Included | Included | Included |
Dated reference. Prices and inclusions can change; check the current pricing page before choosing a tier. All prices are exclusive of VAT.
Frequently asked questions
Do I need an account to scrape Redfin?
No external account is required. You sign up for an API key, receive a starting credit allowance, and send the URL; the service handles rendering and proxies. The current size of the free allowance is listed on the pricing page.
Is scraping Redfin legal?
Public-facing pages can usually be collected, but review the target site's terms of service before deploying and avoid post-login content. Legal advice should come from a qualified professional for your jurisdiction.
What data can I extract from Redfin?
Whatever the public page displays. To get structured JSON back, either supply your own CSS selectors through extract_rules, or describe the fields you want in plain language and let ai_extract_rules locate them. Typical targets are address, list price, sold price, Redfin Estimate, beds, baths, square footage, days on market and the image gallery.
How do I start scraping Redfin?
Sign up, take your API key, and send a GET request with the target URL. The service renders the page, rotates proxies, and returns the data. A free credit allowance is available for testing, with the current amount shown on the pricing page.
Which programming languages can I use to scrape Redfin?
Any. SDKs are published for Python, Node.js, Go, PHP, Ruby and Java, and plain cURL works too. Use whatever your stack already uses.
How much does it cost to scrape Redfin?
A free credit allowance is available on signup, and the current amount is stated on the pricing page. Paid plans start at a monthly fee for higher volume and concurrency. Credit cost per request depends on options: 5 credits for JS rendering, 25 for premium_proxy with JS, and 75 for stealth_proxy.
What is the difference between extract_rules and ai_extract_rules?
extract_rules uses CSS or XPath selectors you write, so it is precise and predictable once a schema is stable. ai_extract_rules describes fields in natural language and lets the service locate them, which is faster for prototyping or when page layout shifts. Many teams start with the AI option and move to selectors once the schema settles.
Why is a field missing from the response?
Check the rendered HTML first. If the value is not on the public page, no extraction mode can produce it. If it is present but your output misses it, the issue is in the selectors or the field description, not in rendering or proxies.