And, I think it’s somewhat crummy of any company to not make downgrading a plan work as smoothly as possible.

However, you said this is for your company’s product, something you bring money in with? And it’s something that you were planning to migrate away from, but hadn’t yet?

I just would say that one - on the other side, why should their business support a paid product for free? I think it’s nice they do, but they could as easily not allow the free plan to be projects that make money.

Also, just as a business operations perspective, planning to migrate and migrating/migrated are very different things.

Also, $99 a year is cheap as a business expense if it’s bringing you in money.

Our fixed costs right now between tech, utilities, etc etc before come in at $4,500 ish per month. And I think we’re very frugal.

Understand different scales but…

Maybe I’m not understanding and I’m being overly critical? If so I apologize.

From a business perspective, I think it’s not wise to treat long time customers that badly, I will never ever come back.

From a founder/builder perspective, I mostly criticise myself: the first subscription was justified, because the license for the icons themselves is perpetual and they are really nice imo. The mistake was to rely on their kits because they are so convenient - of course, 99$ seems cheap at first, but so is this and that and a server here and a paid SaaS there… Sometimes you need to clean up a bit :) the real mistake, however, was to allow the dependency on a provider for such a thing like icons. It would have been very easy to either 1. Self host them 2. Download them at build time instead of run time 3. Implement fallbacks if their servers go down.

Indeed. They could have paid for another 15 years of the service for just what it cost this one engineer to clean up from this mistake. Probably another 30 years worth to stand up and migrate to their own hosted version, assuming it never needs any maintenance whatsoever.

Hopefully they’ll have the sense to not touch it further.

2) The problem, again, is the false claim during downgrade.

3) self hosting a couple of icons is a matter of 15-30min - make a request to the kit, store assets, commit, push, deploy demo, test, deploy prod, done. But not if the kit is not accessible anymore - even though advertised as such.

4) I agree it has been a mistake to ever rely on a 3rd party to host icons. From todays POV, it’s ridiculous I’ve ever made that lazy decision.

When I’m bootstrapping, I tend to value my time at my normal consulting rate, since that’s what we’re ultimately comparing to. If I wasn’t putting this hour into the client on a work day, I’d be billing it to a client, so that’s the opportunity cost in my head.

But yeah, that certainly changes the math a bit. My reply above was in the more normal context of a salaried engineer deciding to save the company £10/month by spending £15,000 in billable hours building an alternative.

I’ll chat with the team and see if we can prioritize a long term solution (we just recently had new ideas for how to do this more simply). In the meantime, we might be able to figure something out for you. Feel free to hit me up directly if you need.