So, here's my experience
My cube is about 25% custom cards, but for the most part (with a few specific exceptions) most cards aren't
that out there. A lot of them are slight edits of existing cards, or ideas close to existing cards.
People certainly come in worried about specific things:
Custom cards they "need to play around or lose". A zero mana fog, or extreme colorshifts like a green
settle the wreckage feel like bullshit if you don't even consider the possibility of them existing. Something to be aware of, but my predilection for fair non-combo magic means a lot of the cards I'm interested in making don't fall into this category.
Power outliers: these are an inevitable part of making cards; you're not going to get everything right on the first try. As long as you take feedback and iterate your designs, they'll land somewhere people end up being fine with.
A note here: it's much easier to adjust cards down than up. In the dark, start with a more powerful version of a card, you'll get more playtest data that way.
The last piece of advice is know what your goals are. I've had a lot of people give me feedback about improving a card in a way I was actively trying to avoid (Eg: this would make it a much better combo card, or why have you blocked off this soft lock, etc), or expanding archetypes into colors I'm not trying for. Check that you're making cards for a reason. (And remember that "This is too cool to not make" is a valid reason too.)