If your app is impossible to scale, then your pricing model isn't aggressive enough. You should have a proper margin to grow your business b/c it'll ultimately benefit your paying customers.
If it's not, someone else would happily figure out a way to take care of your customers.
It's not a pricing issue, it's a time issue. Scaling a production application like Baremetrics isn't a turn-key thing. They can't just up the number of dynos or whatever you're thinking it's like. It takes time, time they decided was better spent servicing paying customers and new trial customers, instead of customers that statistically would never pay them.
I'm not sure what we're disagreeing on, really. The blog post literally has "failure" in the subject. It was an experiment that didn't turn out well, thus they ended it and are now moving on.
If it's not, someone else would happily figure out a way to take care of your customers.