Cloud Cost Savings And The Pay As You Go Model

The Double-edged sword that is the Pay-as-you-go Model

We always open up with something like “Cloud Computing offers cost savings because you pay-as-you-go and because of economies of scale” or “Cloud Computing offers elasticity and seemingly infinite computing resource as long as your wallet is also infinitely flowing” in our write-ups and articles. I know I just did. This is because one of the unspoken tenets of Cloud Computing is “minimal barrier to entry and economic scaling”, so it is ingrained into everyone who knows about Cloud Computing that the main reason for moving to the Cloud is cost savings. But the pay-as-you-go model is a double-edged sword, it can also hurt you.

We call Cloud Computing as utility computing where it is treated like electricity or water, you only pay for what you use. But here is where the similarity ends, water and electricity use are not particularly variable, they are often stable. They very seldom get usage spikes even when there are special occasions in your house. So the utility bills that arrive every month will look fairly similar, only varying slightly. But the cloud is seldom predictable as usage will change depending on the time of day or external events that happen across the globe, and, so usage spikes become unpredictable.

The Cloud usage could just easily increase as quickly as it falls. And this is where the utility feature bites back; it’s only as cheap as its users are responsible. Misuse could easily flush cost savings down the drain. When resources are being provisioned by the provider for the client, billing starts, and it is the responsibility of the client to “deprovision” the resources when, not in use, or set up automatic tools that deal with this. When tenants or users are not responsible for keeping costs down in a pay-as-you-go model, it could easily translate into a pay-more model.

Because of this risk, providers are considering offering their services in a tiered pricing much like mobile data/airtime plans. This is really bad for Cloud Computing as it could be disastrous when the usage spikes and you suddenly run out of allocated resources or time. If the provider allows for overuse and overcharge beyond the allocated limits then it could become more expensive because overuse is often charged more, and it actually defeats the purpose of a tiered pricing scheme.

So as with everything else in the world that needs to be paid per use like water, electricity, gas and food, responsibility is very important in keeping costs down. So make sure that usage is justified so that you only pay for what is actually needed and less on what is used.

By Abdul Salam


Abdul Salam is IT professional and an accomplished technical writer with CloudTweaks. He earned his undergraduate degree in Information Technology followed by a postgraduate degree in Business Informatics. Abdul possess over 3 years’ experience in technical & business writing with deep knowledge in Cloud Computing, VMware,Oracle, Oracle ERP, Cloud ERP, Microsoft Technologies and Network Communications (Cisco, Juniper). Visit his LinkedIn profile at:

Add Comment Here