How do you notice when a data source breaks quietly instead of loudly?

Half of what I ship is a crawler, and the failures I dread are the quiet ones. A company moved its careers page to a JS widget last month, our parser kept returning HTTP 200 with an empty list, and from the outside it simply looked like they had stopped hiring. Nothing alerted, because technically nothing had failed.

I ended up alerting on the shape of the data instead of the status code: if a source drops more than half its rows overnight, a human looks at it before anything downstream moves. That threshold is a guess, and it still misses the slow rot, where a site changes one selector and you lose a third of the fields without ever losing the page.

For anyone running scrapers, ingestion jobs or any pipeline over data you do not control: what actually tells you a source went bad, and how do you separate a real change in the world from your own parser breaking?