Let me pick the one worth telling: on an e-commerce client entering a new market, we carried the creative and the offer over from the home market unchanged. It had been working for months and it looked sensible. Two weeks in we had clicks and no sales.
What had happened
The product was the same; the reason to buy was not. In the home market it was bought as a gift; in the new market the same product competed in an everyday-use category. The ad copy spoke gift language while the price was being compared against everyday-use alternatives. The two cancelled each other out.
On top of that sat a shipping threshold: shipping that was free at home appeared in the basket in the new market and created a loss right at the conversion point. Nothing looked wrong on the ad side, because nothing was wrong on the ad side.
What delayed me
The honest part: I spent two weeks looking for the problem inside the ad account. I changed bidding, narrowed audiences, added creative variants. All reasonable moves, none of which helped, because I was working at the wrong layer. When I finally looked at basket abandonment and competitor pricing pages, it took ten minutes to see.
The lesson was not "localise your creative" — that was already known. The lesson was: when a channel shows a drop you cannot explain, stop looking for the answer inside that channel.
What changed afterwards
A short market-entry checklist came out of it: what is the reason to buy here, what alternative is the price compared against, what do shipping and returns show in the basket, are payment methods local, and in which category do competitors position the same product. These are not advertising questions, but their answers decide the advertising result.
And one measurement habit: in week one of a new market the goal is not sales but being able to see which step of the funnel we are losing. I no longer launch without a setup that shows me that.