Agreed, Nothing worse that a major failure that you know you cant fix but a support engineer can help with ( even if its a case of acknowledging that its being looked into )

Maybe offer a "critical credit" system.. Ie: say, 5 critical credits per year.. You can only mark as critical 5 times.
HOWEVER, if the critical ticket you raised is deemed as critical ( Ie: you were not abusing the system ) you would get your credit back.

At least this way it would stop abusers? but still help those of us who may actually need instant help. ( Storage failure? Network failure? DC offline? etc etc )

Agree, triage is very important. I do not like the idea that I might be standing in the way of a critical outage for another customer with my non-critical question (or vice-versa). I do also agree with a critical credit system to only allow a limited number of critical incidents. DO can make exceptions as needed, but for the most part credits could work. If a customer has more than a handful of issues in a year then the issue is probably them, not Digital Ocean. Probably should issue numbers of credits based on average number of virtual servers or bill amount, so a customer with 10 virtual servers gets a few more credits than someone with one.

100% Agree with this, an 'abuse it or lose it' policy would work great, sometimes you need critical support.

Or implement a "use it for something unimportant and your ticket gets pushed back" system. That way people can always use "Critical" and will get the idea that don't mark critical on something that is not critical to you.

If server is down and user submits ticket they should be moved to the TOP. No critical ticket should be sitting with generic questions. That is ludicrous. Code it so people cant abuse it. If they abuse it give them a friendly reminder of the policy. Three strikes and they are moved to normal queue no matter what.