As an ISV there are several options when it comes to choice of platform. The developers need to be in favor of the choice or you will get into trouble for sure.

But it has to make sense from an economic standpoint as well. From a high level perspective there are three things that drive cost. License feed, hardware/hosting services, and last but definitely not least the labor cost for the whole development process.

Open source of course takes care of the license fees even though you might end up paying for support just the same.

If you choose a commercial platform you are stuck with substantial license fees in one form or the other. It doesn’t matter what the model look like, you have to pay and it will eat into you profits. The question is how much.

If your platform of choice is inefficient in runtime, that is when your customer run your software, it will require a lot of hardware. No matter if it is in the cloud or on-premises. It will drive higher cost. That cost will have to be taken by your customers and TCO for your software will be high. Not good. If you can make this cost variable you are of course in better shape. But lower cost is always better.

Licenses and hardware aside, your main fixed cost driver is labor cost for manning the whole development process. If you can make this process efficient from end to end you are much more likely to be in the black. There are of course other positivt side effects of having an efficient development process. Things like being able to meet new demand from you customers and prospects and being able to adapt to market trends.

In conclusion you need to on top of these three things when choosing you platform. That is what we had in mind when we created Starcounter.

TCO Microsoft MySQL Starcounter