Why is it so hard to fix your processes instead of relying on the HN Support Forum for the privileged few that know it’s an option?

Because at $5 MRR any time Stripe spends on this translates into a loss for them. But it's good PR so they should solve it anyway. Realize that operating a business like Stripe at scale is a lot harder than it may seem to be on the outside.

Today its $5 MRR... tomorrow it could turn into millions annually. They are shooting themselves in the foot. They live to give startups a leg up in terms of monetizing. I doubt this issue would take more than an hour for someone to look at it and realize someone goofed internally. Hell, someone internally goofed and wasted more money probably by not being thorough thereby forcing further time wasted.

Stripe approves accounts without underwriting.. They wait until the account begins to process transactions before underwriting.. This is the result.

I don't believe this to be true. In the US you have KYC laws which are a part of underwriting. We also use Stripe Connect which also onboards new stripe accounts and can watch the automated underwriting happen.

KYC is part of the underwriting process... You have to first know your customer before you can understand their business and assume the risk of processing their transactions..

The catch-22 here is that it doesn't become good PR for them until the customer posts a complaint on social media. If Stripe solved it privately and then released a press release, people would accuse them of either lying or being manipulative, and there goes their good PR.

Yes kind of, although if they fix it they could become the de-facto home for startups. That kind of market ubiquity is hugely valuable. Every time something like OP happens it pushes people (like me) away from using Stripe. I don't want to spend time integrating with a service that might pull the plug from me, and I don't have time to research all the various little ways that my business might be "too risky" to extreme regulation. I get that some stuff is too risky, but a service that lets me talk to a human before wiping me out would be real nice.

HN Support = press prevention plan

What's the maximum number of false positives they are allowed to have before you consider the process "fixed"? And do you think it would be an improvement if they purposefully didn't check HN for bad support experiences?

It'll be fixed when the number of "but they denied my appeal with a short template response" drops to 0. A template response means they didn't even read the appeal, which has nothing to do with the original false positive.