Web Scraping is a viable option to keep track of real estate listings available for sellers and agents. Being in possession of extracted information from real estate sites such as Zillow.com can help adjust prices of listings on your site or help you create a database for your business.
In this tutorial, we will scrape Zillow.com, an online real estate database to extract real estate listings available. This real estate scraper will extract details of property listings based on zip code.
Here are the following details we will be extracting:
- Title
- Street Name
- City
- State
- Zip Code
- Price
- Facts and Features
- Real Estate Provider
- URL
Below is a screenshot of some of the data fields we will be extracting
Scraping Logic
- Construct the URL of the search results page from Zillow. For example, here is the one for Boston- https://www.zillow.com/homes/02126_rb/. We’ll have to create this URL manually to scrape results from that page.
- Download HTML of the search result page using Python Requests – Quite easy, once you have the URL. We use python requests to download the entire HTML of this page.
- Parse the page using LXML – LXML lets you navigate the HTML Tree Structure using Xpaths. We have predefined the XPaths for the details we need in the code.
- Save the data to a CSV file.
Required Tools
Install Python 3 and Pip
Here is a guide to install Python 3 in Linux – http://docs.python-guide.org/en/latest/starting/install3/linux/
Mac Users can follow this guide – http://docs.python-guide.org/en/latest/starting/install3/osx/
Windows Users go here – https://www.scrapehero.com/how-to-install-python3-in-windows-10/
Packages
For this web scraping tutorial using Python 3, we will need some packages for downloading and parsing the HTML. Below are the package requirements:
- PIP to install the following packages in Python (https://pip.pypa.io/en/stable/installing/ )
- Python Requests, to make requests and download the HTML content of the pages ( http://docs.python-requests.org/en/master/user/install/).
- Python LXML, for parsing the HTML Tree Structure using Xpaths ( Learn how to install that here – http://lxml.de/installation.html )
The Code
You can download the code from the link here https://gist.github.com/scrapehero/5f51f344d68cf2c022eb2d23a2f1cf95 if the embed does not work.
If you would like the code in Python 2.7, you can check out the link at https://gist.github.com/scrapehero/2dd61d0f1bd5222a4c9ae76465990cbd
Running the Scraper
Assume the script is named, zillow.py. When you type in the script name in a command prompt or terminal with a -h
|
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 |
usage: zillow.py [-h] zipcode sort positional arguments: zipcode sort available sort orders are : newest : Latest property details cheapest : Properties with cheapest price optional arguments: -h, --help show this help message and exit |
|
1 |
python3 zillow.py 02126 newest |
You can download the code at https://gist.github.com/scrapehero/5f51f344d68cf2c022eb2d23a2f1cf95
Known Limitations
This script should be able to scrape real estate listings of most zipcodes provided. If you would like to scrape the details of thousands of pages, you should read Scalable do-it-yourself scraping – How to build and run scrapers on a large scale and How to prevent getting blacklisted while scraping.
If you need some professional help with scraping complex websites, you can fill up the form below.
You can also get data delivered to you, as a Service from us. Interested?
Turn websites into meaningful and structured data through our web data extraction service
15 comments on “How to Scrape Real Estate Listings from Zillow.com using Python and LXML”
Excellent tutorial! Is it possible to have the code search individual properties? For example if I had a list of 600 addresses in excel, could we perform the scrape in this article on each of these properties? Can it be modified to use the specific address as a dynamic user input into the website and then automatically return each property in order?
Thank you in advance for your time.
Hi Alex,
Thanks for your feedback.
The tutorials provide a starting point for your own specific use cases.
Yes you should be able to do what you need by using excel macros and building a a scraping api endpoint.
Thanks
Great script! And very easy to use. However, it only reaches the first page of the Zillow results — in other words, 25 houses. Is there a way to get it to pull all houses available in the zipcode searched?
Hi Howard – that is an exercise for the reader – a key point of the tutorial. Have find coding !
Nice article, it seems zillow will only show 500 (25/page * 20 pages) data, in order to fetch all datas in some super hot zipcodes. We might need to add filters to the search which I think price range (0-10k, 10k-20k) might be a useful attribute to narrow down the data size. The problem with price range is that it hugely depend on the area, by which I mean if some area only have houses price from 3m – 4m we will go through a bunch of unnecessary requests and this will increase the probability to get blocked. What do you think is the best attribute to limit results within 500?
Ran the script, but it gets detected and Zillow is passing back a captcha now. “Please verify you’re a human to continue”
You could use some techniques listed here to get around it – https://www.scrapehero.com/how-to-prevent-getting-blacklisted-while-scraping/
was anyone able to get around the “Please verify you’re a human to continue”?
Awesome article. When attempting to run it on the agents page it returns empty string using multiple classes. In example this page: https://www.zillow.com/malibu-ca/real-estate-agent-reviews/?sortBy=None&page=3&showAdvancedItems=False®ionID=12520&locationText=Malibu%20CA
I see there is another head and body loading at the end of the page is this a trick to prevent it from being scraped? Do you have any work around to this?
Nice scraping logic. A few tweaks and you can use the code on other real estate sites. Thanks!
If scraping Zillow page, does output CSV capture the property feature codes (eg a RESO compliant set of listing features and description)
just dont get it. why my csv was empty but fieldnames.
I’m also getting empty csv
hi, same here, can’t get any csv output beyond the defined headers!
what to do?
thanks,
kurt
also, running on win 7, python 3.6.4