In practice people don't think about or even look at URLs, that's why phishing works. They recognize sites based on logos and stylesheets and often go to places by googling the brand name and clicking on the first link. If you forward every domain to a primary domain at the DNS level, your pagerank probably won't be hurt by this practice.
to go a step further, adding out of band contact info is a big deal.
Forward mycompanyname@gmail.com to info@mycompanyname.com and give it out in the emergency support info for support contracts Add it to your status page as needed, (which should be running on someone else's service or mycompanynamestaus.com).
Customers that really need your service, like the ones who pay, will check the status page and can update the endpoint as needed.
Make sure your sip lines don't point to mycompanyname.com.
If you publish a client side app, use 2 domain names as endpoints, mycompanyname.com and mycompanyname.io. Have the app or service check for and fail over if one doesn't work.
Make sure paging and technician notification is handled by a system that won't be affected by this. (nothing more amazing then getting 200 pages AFTER you've spent 2 days recovering a total failure of a system. You just want to go to sleep but you have to wait for the email queue to drain since you can't turn your pager off.)
Either way, use 2 domain names, and set them to expire at 6 MO intervals. Buy the domain for 2 years (or more) and renew every year so you always have 1-2 year lead time to sort out issues.
The list above would probably cost about $200/year and a few extra hours but it keeps you from getting backed into a corner. Everything else in our infrastructures has fail overs, and limited blast radius for failures.
We tend to us domain registration as a single point of failure and one one even things about it.
