I know about different bar - not in customer service business, in software engineering. You're given a task to write a parser in 15 minutes - literally, you solve it, it mostly works - for some simple inputs at least, you don't have time to test it thoroughly. You don't pass - the bar is different.

Your experience tells that the startup is too early to think about scalability problems. They can keep things simple for a solid another iteration of the product. You explain that to them, they listen, but they are doubtful - they've heard that scaling is crucial, and even though they have like a hundred of customers, they want that - so your advice is considered wrong. You don't pass that bar.

They tell you they need somebody senior to coach and mentor a bunch of more junior engineers, somebody who'd steer the whole technical stack. In remaining minutes of the engineering interview they ask you to write Rabin-Karp implementation, or AVL balancing. You don't pass the bar if algorithms don't work first time.