8 comments

[ 3.4 ms ] story [ 32.7 ms ] thread
Is there any cleaner data source for home listings? I wish I could just setup a system to automatically filter listings by criteria that are important to me.
Unfortunately no, the underlying systems companies such as Zillow use require basically joining to access to the local Multiple Listing Services (MLS) of a certain market. All of these have different rules of how data can be used.
I tried to scrape zillow in past (2019) and it was giving me captcha headache. Didn't author faced similar issues?
> As of 2022, Zillow uses PerimeterX for scraping countermeasures. This can be bypassed by using ISP proxies from BrightData.

Why ask without reading the article?

Well, I used luminati( Currently BrightData). And, it was still giving me captcha. In fact, it was giving me more captcha in BrightData. So, I am not sure.
There are services that use humans to solve captchas for a very reasonable price, such as 2Captcha.

Some of the protections are better than others. For example, you may need to use a real browser with Java script support instead of just HTTP requests.

Browserless works well if you enable the stealth functions.

what’s the point of scraping a scraper?
You want a machine-readable dataset rather than a website, and the aggregator has already done some of the work for you.

Scraping all of Zillow's sources instead would explode the complexity of the project.