Case study — fig. 03
Ownership was being assigned by memory and tribal rules: who covers what, who's out this week, which accounts moved when someone left. Every mis-routed record was a delayed response wearing someone else's name — and in a busy org, mis-routing isn't an event, it's a background hum.
A territory assignment engine, built to live inside an org that already ran 130 flows — which is its own discipline. The job wasn't just "make assignment work"; it was "make assignment work without disturbing the other 129 things this org does every day." The engine was covered end-to-end with Apex tests before it touched a single live record, so the rollout was a verification, not a gamble.
1,500 records re-routed across six owners in one pass. Zero failures. Assignment now happens by rule, not by memory — and when the rules change, the tests say so before production does.
In a mature org, the hard part is never the feature. It's adding the feature without breaking the hundred things nobody documented. Tests aren't overhead; they're how you move fast inside someone else's decade of decisions.
Running an org where every change feels like surgery? Tell me where work is falling through — I'll tell you the first three places I'd look. Free, no pitch.
Find the leak or just message me →