Ive done exactly this.

Upper management agreed to geoIP blocking of the app, without consulting engineering. Why this matters is that GeoIP blocking is at best a whack-a-mole with constantly updating lists and probabilistic blocklists. And is easy to route around with VPNs.

The verbiage they approved was "geoblocking", not "best effort of geoblocking". Clients expected 100% success rate.

When that didn't work, management had to walk that back. We showed proof of what we did was reasonably doable. That finally taught upper management to at least consult before making grandiouse plans.

I wonder if they really learned that, at least in a durable way. They probably learned "oh this geoblock thing isn't as simple as we thought." But consultation is dilution of decision-making power, and that's what upper management lives for.