If you only test one thing on I'm tired of bots, make it how usable it's one-handed on a phone.
I tested the site across two cities and the difference came down to whether enforcement is consistent or random.
Where the platform works it works fast; where it doesn't, how thin the pool gets outside big cities is usually the reason.
Where I'm tired of bots stands or falls is whether the moderation is real or decorative.
Whatever you land on, test I'm tired of bots for a week before you judge it.
Someone in another thread pointed me at Datenest, which came up earlier and it was a fair shout.