For what it’s worth, I think $4/month is absolutely affordable, and very fair. In fact, it’s at the perfect sweet spot for me where it’s a no-brainer (I start to get cheap at $5/mo and up, and especially after around $8/mo). And for those existing users who are feeling a bit “salty”, I’d urge to think be grateful for the time we’ve had Beeminder for free and consider how in practice that means we’re getting a huge discount and really paying something like say $2/mo. You’re getting an even better deal!
I think $16 and $32/month is a bit steep, but I’m not sure how many users will actually be using these plans (you mentioned 28 people are currently on $32/mo Beemium, do we have stats for the other tiers?). As it stands now, I don’t really need the Bee Plus power features, though there’s one goal of mine that I’d consider paying a one-time fee to enable autoratchet for. The problem is that it’s not enough to justify paying quadruple the money for that privilege.
I wonder if you might make more money by dropping the power user plan to $8/mo (I derived this from gut feeling, but it seems to match some others in this thread), or by allowing users to pay a la carte for specific goals (either a one-time payment or a slightly higher monthly subscription rate). Probably not if you get sizeable numbers subscribing to the $16/mo plan, but worth considering if almost everybody stays on the $4/mo plan.
As for the $32/month tier, I feel like this is the plan for people with a lot of money to spend and/or want a way to support the Beeminder developers. In terms of features, I feel like this and the $16/mo tier could be merged, but I understand why for psychology reasons you want to have three tiers, so I think that’s fine.
I do think that maybe for newbees, a time-limited trial with unlimited goals (both number and type, though I understand the concerns about some types being confusing) may make more sense. Yes, it might feel like you’re holding user’s additional goals ransom, but it should be fine as long as it’s extremely clear the entire time. For example, in the dashboard/gallery view, you can have a dedicated section for your three “free” goals, and then a separate section for the rest (either “premium goals” or “temporary goals” or something), and let users drag-and-drop goals from one section to the other.
One potential issue is that, depending on a user’s goals, a free trial may take a relatively long time for users to fully grok Beeminder, so I can see an argument for a free trial that lasts say 3 months. Forgoing those months of revenue may be well worth it though if at the end of the day you get a Beeminder convert who will pay $4/mo for several years, vs. trying to monetize users too quickly and only getting them for a few months or losing them altogether.