Why one client's problem becomes everyone's problem
Two things make this spread: a shared sending domain or IP pool, where one client's poor list hygiene drags down a reputation other clients rely on too, and simply not knowing which client's domain is the one causing a problem until an overall number looks bad. Isolating sending per client, and checking each domain's own posture rather than an aggregate, is what keeps one bad list from becoming every client's deliverability issue.
Check this before you take on a new client's sending
Before you send a single campaign on a new client's behalf: verify their existing list (a purchased or long-unused list is the single most common cause of an immediate bounce spike), and check their domain's current SPF, DKIM and DMARC posture rather than assuming it's set up correctly. A domain with no authentication at all is a liability the moment you start sending real volume through it.
Keep checking, not just once
A client's domain posture and list quality aren't fixed facts - a client might add a new tool that sends as their domain without telling you, or a once-clean list might go stale over months of no sending. Re-check each client's domain on a schedule, and verify lists again before any campaign that's been sitting unused for a while.