I wrote a script in 2018 to go to zillow to check my house values every week:

image credit: bestproxyreviews

• they announced that they would stop the free API 2 years ago
• they actually stopped the access 1 year ago, forcing everyone to scrap or pay for a high touch corporate service
• they changed twice their pages layout since then, but nothing that small adaptations of the XPath could not do
• it appears now that the page retrieved by simple request is different from that with selenium

The page is now fully dynamic, and more expensive to retrieve automatically.

My guess is that many of these commercial scappers might be broken.