A residential proxy setup can look “fine” in testing and still fail in production.

A 200 response does not tell you:

• whether the page is challenged
• whether the geo is correct
• whether retries are already too expensive
• whether the session survives long enough for real workflows

For scraping, I think “cost per successful usable page” is often a better metric than cost per GB.

What are you measuring first when validating a proxy pool?