It is unacceptable to know of a multi-hour outage to an application people feel is critical, and not post an update. We're not looking for a detailed analysis, just something to say, "Whoa, as most of you have noticed, something broke. I'm working on fixing it and will give you a short progress report every two hours." It is reasonable to expect a 30 second post every two hours to the people who are your paying customers.
This is what a professional organization that cares would do. Whereas the service level I am observing is more akin to an organization that has their servers hosted in someone's parents basement and the sole admin is passed out drunk from a night of partying.
I'm not trying to beat anyone down, but your outage procedures definitely need revision.