Subscriptions. If you're going to keep people permanently on staff, a steady revenue stream is required. Either the users will provide that (aligning the incentives of the company with the users, since financial performance directly correlates to customer satisfaction), or other companies will, in the form of ad spend, corporate partnerships, cross-promotion, whatever.
Possibly the more sustainable model overall is for software to become a gig job, where engineers get paid for the features they deliver and there's only a skeleton crew of maintainers keeping the project on track (to keep the ongoing costs minimal).
To offset the cost to users even further, you could potentially apply for government grants, at least in Europe. Or corporate sponsorships in some places, in the vein of the Django Software Foundation.
Firefox and Signal are both reliant on market share to stay relevant, and I would say they cannot afford losing users to a subscription model.
It could work if the base product was free and some addon could be purchased as a subscription, but that is also pretty much happening already with Mozilla VPN.
It is all over the comments here: People want to donate. They want to donate to specific products they like though, and not to the organization in general. Donations need to have a specific goal, like for example further development of Firefox and keeping it clean of stuff people do not want. When donations are not specific enough, like it seems to be the case with Mozilla, they run risk of being misused. Then the actual thing people want to support are not actually supported by their donation.
Signal is run by a known angry dictator. The same goes for Mozilla. These companies don't mean well for users, they mean well for themselves. The rest is the narrative sauce that they abuse to insert themselves into a specific niche.
