No, web is not the last refuge of software, and $20/mo. is not a high price point.
Before my current job I had no idea of the sort of money that corporations would throw at enterprise software if it filled an need.
We have an enterprise product in a under-developed market; in the first six months we pushed prices up literally 1000 fold without any drop in interest.
I've observed this even in competitive markets.
Enterprise is a seriously seriously big money spinner; you're looking at more cash in a single sale than some of the exits we see.
EDIT: and the irony is that most of this software is not particularly hard to build. A lot of it is just databases with a front end on it; the only bar to entry is being able to work your way through the policy document/guideline/law that you are providing a tool for and figure out what is needed. After struggling to throw together relatively complex web apps for a couple of years, this sort of development is almost cheating :D
"An operating system," replied the programmer.
The warlord uttered an exclamation of disbelief. "Surely an accounting package is trivial next to the complexity of an operating system," he said.
"Not so," said the programmer, "when designing an accounting package, the programmer operates as a mediator between people having different ideas: how it must operate, how its reports must appear, and how it must conform to the tax laws. By contrast, an operating system is not limited by outside appearances. When designing an operating system, the programmer seeks the simplest harmony between machine and ideas. This is why an operating system is easier to design."
- from the Tao of Programming
Yes, I didn't believe it either.
And this is just one of the many line items.
According to VC, some of our bigger customers didn't take us seriously until we charged the money they were used to paying.
The problem with building enterprise (highly-domain specific business) software is that to start, you need to know that a problem exists.
You won't find this information on the Internet or even in books. You need someone with very deep domain knowledge and the insight to recognize that a problem exists, that it has a possible solution, and that the solution has a larger market it can be sold to. The people who know all that, and are willing to take a risk and start a company around it are rare. If you're a programmer, find and work with them.
Figure out a way for a company to do most of X itself with some software whilst still leaving some "wrapping up" or auditing for consultants to do.
Convince consultant to come onboard with their expertise to help develop the software (revenue sharing deal). He/she becomes a ready made sales person once launched.
Price product at fraction of the cost of getting consultant to do the work (i.e. instead of 100K - price it at 10K).
Any tips on the "find a consultant who charges a lot of money to go and do X at a company" stage? Do consultants have conventions or anything like that?
Most web companies would kill (but not alas sell their startup soul) for the average 10 - 15% of the initial license fee for enterprise software that big companies pay annually for ongoing support, maintenance and upgrades. Similarly, I just heard of another iPhone app that clears $6,000 a month. And it's not even at the top of its list when you search by the niche keyword. Some of the decent cash-flow being made in the App Store is never blogged about.
The six percent sales commission that top enterprise salespersons can get on a seven figure deal can easily eclipse a startup's lifetime revenues. Enterprise software is hard because it deals with real world complexity. Consider a payroll system. How do you deal with exceptions? The employees who are grandfathered and get certain perks. The Quasi-legal employment contracts.
There is a famous saying that if you buy SAP software you fit your company around the software. That is you adjust your internal processes and controls to meet The One True SAP Way.
Seriously, I don't understand that part. I'm not sure I understand your main point either, come to think of it. Non-web software matters? I don't think that was disputed.
Web software doesn't deserve the focus it gets on HN because it almost always starts out sans business model, and then half of the population never figures it out, and the other half does, and of that half, 2/3 fail to execute? If anything, that's a description of a pretty exciting, if somewhat high-variance, domain.
And wish your friend much good luck for when - not if - he will try to find a job again.
Funny point: the non-web companies I do freelancing for are totally ok with me using MongoDB, Ruby, JRuby or whatever does the job. They mostly don't care (in my case at least).
There are a few glaring exceptions like Google, but for the most part conventional desktop software sells much better than web software. Many web startups have no real monetization plan and are going nowhere fast, some inspired by attitude embodied in The Social Network's telling of "I don't even know what it is yet, how can we monetize!" I've actually had a friend tell me that his startup is not going to do anything to make money at first, because "like Facebook", they "just want to see where things go", or something like that. I think this attitude is pretty dangerous and misguided.