A few billion people who currently have no bank accounts could suddenly participate in global trade.
I also doubt that people who don't have bank accounts have private computers. So really the pool is "people with computers and bank accounts in countries so replete with fraud that Paypal and CC companies don't service them".
The argument is that they'd be paid in Bitcoin for goods/services in the first place. It is a currency, after all, just (perhaps currently, perhaps always) a very volatile one.
Bitcoin beats "Cash On Delivery" because it is trustless, but that's easy to address by having the delivery service allow a remit prior to the shipment.
Note that with CBD, a (cash using) recipient has to deal with 1 party, with bitcoin they have to deal with 2 or 3 (obtain, spend, receive).
Services would be somewhat different, but it still seems like quite an open question whether bitcoin would (in practice) be better than buying some credit from a payment provider (Paypal or Amazon or whatever).
But you could have said the same thing about email/the web in the past.
A global communication network and protocol without anyone using it is pointless. The same for a global value transfer protocol. It seems obvious _a_ protocol, probably blockchain based, will gain mass adoption eventually.