I am the author of the article.
I agree that time is valuable, but if you charge yourself out at, say, £600 a day, and you save 4 grand building a phone switch in 2 days, it's an easy decision to make. It wasn't time squandered. I save money every time I need to make a change to the phone list, as I dont have to call out an engineer.
If you are a 2 person company, then there's obviously no point. We built our phone switch when we moved into our own offices and hired employee number 5, which I think was about the right time to do it.
The company I work for scaled from two people on the queues to 15+ desks+cells and counting -- 50 employees to 1500+ with only the most minor tweaking. For a while we also had IAX trunk links to major vendors and other companies that we spent a lot of phone time with.
I can't recommend it enough.
TWO DAYS of hacking spent (squandered?) building your own phone system? Really? Unless you have significant staff, how can that be remotely a reasonable return vs just using cell phones?
Many of the other suggestions are good ones, but don't get caught in the trap of "Just because I can do XXX which will save $Y, that I obviously should." If your core competency has something to do with phones, by all means hack away on your phone system.