The pricing issue, of course, is often at the core of the decision model used to decide to go into the cloud. So couple of things are important here. First of all, a lot of people, of course, have as an initial motivation to go to the cloud, that they believe it's going to be cheaper to operate in the cloud than to operate on-premises. It fairly often ends up being true, but not necessarily. So you have to understand why it might be, and why it might not be. First of all, I would say, to me and to many types of enterprises and clients and governments that I have had the pleasure of working with, the main financial impact is not the reduction in total cost, if there is one. It is the fact that you're replacing capital expenditures with operating expenses. So instead of shelling $1 million to buy a system, and then having the financial people do the amortization over three years or five years, whatever the law of the country requires. Instead, what you do is you're starting to rent as you go for a certain number of licenses that is supposed to fluctuate according to usage and be proportional to your usage. That in itself can represent a significant financial advantage because of the cash flow aspect, you don't have to come up with the money up front. You're going to only pay progressively as you go. So if you're an IT manager, you should go and talk to the financial director or the CFO, especially if you yourself are not an expert in corporate finance. And you should discuss this and you should work together, side by side, with the CFO's preferred tool, which is Excel, and say let's model how much we're going to have to pay each year under certain usage scenarios, and then the CFO will tell you what makes more sense. So some of the advantages of going to the cloud is that if you have a fairly long ramp up period, like you're doing a pilot project with 10 users and then you're expanding to 100 users, and then you're expanding to a division of the company, and then you're expanding to the whole company. Well, you're not going to have to buy the total system up front, you're going to be able to start ramping up. Which means that if any accident happens in that ramp up and the project is abandoned, the amount of money you will have lost during that ramp up period, that will maybe abruptly end, is going to be much lower than what you would have wasted if you had bought fixed licenses or fixed amount of hardware. Similarly, if you have a downturn in the business, you should expect the cost to go down. If you have a purchase of another company, or you are purchased by another company, and therefore, one of the systems is abandoned, or all the users are coming onto the system you have, etc., the price will modulate according to your usage. So these are the advantages about cost. It's the flexibility, the ramp up, the ramp down, the conversion from capital expenditures to operational expenses. It is not necessarily the total cost of ownership over the total lifespan of the asset. Which means that among the criteria you should consider to move to the cloud or not is how comfortable, how confident are you that you can predict the usage over a fairly long period of time? If you are ready to bet that you're going to use this new capability, for instance, a new application, for five to ten years and you don't see any chance of that plan being derailed. Maybe it is better to buy a solution outright than implement it internally, but if there are lots of factors that might change your usage over time, the flexibility of a cloud rental model is probably to your advantage. Now, the pricing schemes that the vendors offer may or may not really let you implement that flexibility, so that's where you have to pay very, very much attention to the cloud agreements that the vendors are going to ask you to sign. Very often the pricing options that are being offered are reasonable or they offer a certain variety, you can choose from several schemes. And if one vendor doesn't offer you the schemes, the pricing, and invoicing methods that you like, maybe you need to go and look at another vendor. But you really need to look at, how long are you committed a certain level of expenditure for? And if you're usage of the resource, the asset, increases or decreases, how fast, or how slowly does your cost evolve to match your level of usage? You don't want to pay for multiple years at a fixed rate even if your usage decreases by half, you want a very short delay between a decrease in activity and a decrease in costs. The vendor will always be happy to sell you more resources and increase your bill. But are they going to go in the other direction and reduce your bill quickly enough if, for some reason or another, your usage decreases. So that's what we call elasticity in general in the cloud. Like an elastic when you stop pulling on it it becomes smaller again, shorter again. That's what you want about the pricing scheme and you need to examine that very carefully.
