Companies Who Make Money: Virtualmin (YC winter 07)

In all seriousness, though, being the Open Source nerds that we are, we cleverly trick our users into learning lots of interesting stuff about Apache and Linux and other stuff. When things need to be complex (as things occasionally do in a virtual hosting deployment) we don't make up our own complexities and leaky abstractions and slap them on top of the existing ones--we allow the user to see through to the Apache/Postfix/BIND complexities. For example, with redirect rules and match patterns. Very basic match patterns can be setup automagically, with very little effort...but when things get complex, the user can dig down to see the actual Apache rules. Likewise for almost all of the services we manage, if the user needs fine-grained control they can get to it without having to completely break out of the comforting Virtualmin UI. Baby steps to system administration competence.

We also spend a lot of time in our forums talking about the whys and wherefores rather than just the "Click here, click there, and save it".

We genuinely love the software our products manage (mostly). And we've both written books on the topic of system administration, in total ~1100 printed pages, which have now been combined into a massive wiki ( http://doxfer.com/Webmin ) and that's also a pretty good source of traffic and potential buyers for our products. Interestingly, our books are still selling a reasonable amount five years after publication (folks like tangible products, I guess).

We actually view education as a good marketing tool. The more people know about system administration, the more they'll understand why our products are dramatically better than cPanel or Plesk. The first time someone hits the command line with our competitors products they'll learn a hard lesson in the difference between template generated configuration files, and a product that understands the configuration files and respects non-generated directives and comments and file order. People don't go back to other products after spending enough time with Virtualmin and Webmin.

Actually, we have both an Open Source version and a proprietary version. The proprietary version adds a few nice features, primarily in automating installation of about 85 popular web applications, as well as some other features related to making money with virtual hosting systems. And, it has a price, and people pay us money in exchange for the software.

I don't know the author (I don't think, though maybe I've talked to him on our forums or issue tracker; we have 1400 or so paying customers, and many thousands of Virtualmin GPL users, I can't really keep up), but we're happy to have him as a Virtualmin GPL user, regardless of whether money changes hands, and not just because he wrote a really nice blog post about us. We have absolute confidence that if ever he does need the features that Virtualmin Professional offers, he'll buy from us. We've found the single best marketing we can do is to get people to try our Open Source projects...nothing else is as good an indicator of whether they'll eventually buy our product than use of Webmin, Usermin and Virtualmin GPL. Which is good news, since we have several million Webmin users that haven't bought anything from us yet.

So, perhaps the interesting question will be whether we can build a company bigger than Parallels (or even cPanel) on a mostly Open Source basis. We're doing OK, so far, and growing at a nice pace, but nothing anyone would point to as an overnight success. But, we'll see.

Anyway, the author is not mistaken. We do make money, and the amount of money that we make is increasing at a respectable pace. It's enough money that I feel like my initial optimism about the business was well-founded (I feel like I picked the winner from among my stable of ideas) but not enough that I can slack off. So, it's probably just the right amount of money for this company at this stage.