Cheezhead

Via Jim Holincheck, a Gartner analyst in our space, the Top 10 Laws for Saas Firms, by Byron Deeter, Bessemer Venture Partners.

Born to break the Law(s), I have to disagree with at least three and I suggest that the sample from which the rules were derived is not representative of the universe of SaaS providers.

First the rule about no on-premise solutions; that’s going to kill a lot of deals because SaaS is a delivery model, not a business model. If firms already possess quality Internet application infrastructure, database licensing and admin, and advanced BI capabilities, why would they pay margin to SaaS firms ? Why accept integration and extensibility compromises to boot ? Furthermore, having to code, test, version control, and deploy your application tends to enhance your own abilities as a SaaS vendor in terms of CPU/storage/bandwidth resource use, disaster recovery, documentation, and general internal TCO.

Because you are offering on-premise solutions does not automatically mean multiple code-bases; if the design and business model is effective, it’s economical to maintain reasonable backward compatibility and support for recent versions of installed apps. Multi-tenant can actually mean a wide range of choices in solution design- if you don’t want to lop off a third of your market, you better believe that hybrid vendors may have an advantage over time as the reality of fixed and marginal cost infrastructure cases is realized.

Second, as clearly seen by the comments, the law limiting at one data center is probably more a function of sample distortion of the vendors involved- smaller, younger vendors obviously will start with one, but customers, geography, and redundancy needs may indicate two or more in short order. If the application/ solution architecture is not easily amendable to multiple data centers, just how good is it?

Third, the rule about 1M a month in MRR before hitting Asia or Europe is just silly. If the phone rings, and the person on the other end wants to buy, you sell them a subscription. Who cares where they are sitting? If your localization strategy for the app is strong, you may even be able to offer local languages with little increase in your cost basis.

The sample used to derive these rules to me clearly demonstrates the low(ish) barriers to entry for new SaaS firms- all you needs is some code, some servers, a phone, some adwords, and off you go. The big ERP and leading vertical app players did and do have a vested in interest in their old delivery models and channels, but don’t think that as they get into the game for real that the standards won’t rise for all the players- highly robust web services, highly scalable infrastructure, and complex integration tools will become requirements to sell into larger and even medium businesses in a few years. Likewise salesforces, consulting, and technical service people will need to be trained to higher levels and that will cost more. The cost curve for SaaS is unusual- very flat for a few customers, steep as you gain volume, and then flatter when you are pretty big, then steep again when you are huge. Certainly not the old industrial model of lower unit costs with higher volume.

These 10 rules are reflective of a gold rush, and it’s real enough, but as always, the wild landscape will be tamed and the rules for everyday life in the territories will look a little different; and probably a lot more like the civilization left behind than it may appear to people digging like crazy.