This may be the case in some instances, but in most situations I run into, the cost is not the main driver.

Let me agree with you for the sake of the argument that cost could be the driver. I’ve run into many “reports”, but like I said, most of these reports are sponsored by either on-premise or on-demand companies, so it’s hard to take them as facts. I’m not sure if any academic research was done on the topic. I also have not seen any downward pressure on hardware costs, but maybe we will see some of that happening to license costs. This being said, I again doubt that even a substantial decrease in the license and implementation cost would make the balance tip in one direction or the other… for this reason, I doubt that SaaS vendors would have to adjust their cost to make their solution more attractive to people considering an on-premise solution.

The decision often comes down to how the company is most comfortable to pay for the solution, if they have the in-house capabilities and desire to maintain it, how concerned they are about data security (I’m not saying that SaaS is at higher risk), and what has been their past experiences with such a solution… And that’s assuming that the on-demand solution could meet their needs.

Accepting my contention, are there any published reports or empirical client analyses that address the following:

1. The single biggest item making on premise prohibitive?

2. Cost comparisons that assume a decrease in asset price?

3. Tipping point where clients lean back towards on-premise?

I ask b/c the looming depression will significantly affect purchase prices of everything from shampoo to servers to the downside (good for purchasers), and I’m wondering if firms promoting SaaS have conidered any of this in their analysis. There is a point where OnDemand loses it’s cost savings mojo, most would assume it’s below 1:1 but since OnPremise is still the baseline I would surmise it’s well above that but below it’s current point.

kt