Yes, good point.  
Auth + Rate limit = Great. They might be good enough for quite some stuff.
But "Worthiness of a free trial" translates into "Likelyhood of using it paid" real fast when you look at products/services that are social, like many Nostr things inherently are. 

Letting tons of people try out your Tennis club for free might actually be a great idea on a specific "Discount Day" or something like this. But you don't ask your paying club members to invest time and energy into a bunch of new people aaaaaall the time. I've lived that life for a while in an ecovillage and it's bad UX.
My guess is that, just like for those real-life examples, you don't simply let people in without some sort of skin in the game. 

 I think that makes sense for custom or community relays, but the generic social media ones are going to find it tough because they have no target group. They're probably going to have to link relay usage to their client, somehow.

I guess that's why Bluesky went invitation-only. 

 You're right about constant, free trials on public relays being a burden to current customers, tho. Creates npub churn.

You always see that effect when we have the onboarding waves and then everyone emotionally invests in welcoming newvbies and then most of them disappear again. Wears down the welcoming enthusiasm, with time. 

 Y'all KWIM?

We need to get better at targeting and concretely onboarding the high-potential users, or we're going to exhaust and demoralize the welcomers and waste relay and dev resources.

We're too inefficient and the offer isn't differentiated enough. We need more tiered pricing models, for instance, so that people see that they can upscale and get better service, rather than only offering low-scale service to everyone for free.

nostr:nevent1qvzqqqqqqypzphtxf40yq9jr82xdd8cqtts5szqyx5tcndvaukhsvfmduetr85ceqydhwumn8ghj7argv4nx7un9wd6zumn0wd68yvfwvdhk6tcpr3mhxue69uhhg6r9vd5hgctyv4kzumn0wd68yvfwvdhk6tcqypxrkmllsleme3n0wpf6qw62f45x6utl4ehq3apc6w837vmfs6eqxw4n8e4