One of the greatest advantages of cloud computing is its flexible price structure. The idea behind it is simple: Cloud users only pay for what they have used in terms of resources from the cloud. However, this usage-based option is just one of the many options that cloud providers offer their customers. Find out more about the most common pricing models for cloud computing.

Before you start looking for the right pricing model and comparing cloud providers, you first need to analyze your own business. As a rule of thumb, you should find the price model that best suits your business model – not the other way around. It shouldn’t be necessary to change your business to fit the billing system as closely as possible.

If, for example, you are looking for a suitable IaaS solution, you should look carefully at services such as the provision of server, computing and storage capacity as well as traffic, support and data security. With SaaS and PaaS, on the other hand, the software and hardware used, such as Microsoft Office 365 or AppAgile, is of crucial importance. Answer the following question for yourself and your business model: Is flexibility in the pricing structure important to you because you anticipate dynamic loads, or is a long-term model with reserved flat rates more sensible because it provides greater planning security?

But which price models are available? In a study, the analysis company 451 Research attempted to categorize the price models in the field of IaaS and came to the conclusion that the market is very diversified and that it is very difficult to draw transparent price comparisons. In principle, each provider’s pricing structure differs in the details, but there are three basic models used for cloud computing.

The pay-per-use or on-demand pricing model means that you only pay for IT resources that you are actually using or have already used. This option is usually free and easy to sign up for, which means that you can deploy cloud services in minutes, and you have control over which components you purchase from the cloud and ultimately pay for. At the end of the month, each cloud user receives a detailed list of all the services they have used over a given time period.

The pay-per-use model is thus particularly suitable for small companies or start-ups that have difficulty forecasting the development of their own business and depend on flexible structures. For example, if an app developer wants to run an application without needing to invest any money and has a hard time estimating success (load peaks), they will take a low-risk approach with an on-demand model that can flexibly scale resources.

The pay-per-use model offers maximum flexibility due to its scalability, but is not the ideal choice for every business, as the following example illustrates. If a customer wants to run their website in the cloud, they need continuous operation with a minimum configuration. For example, although the least traffic and thus the least main load is to be expected at night, they will avoid restricting the use of resources at this time of day as it could result in conflicts with users from other time zones. For continuous operation of an app, a pay-per-use model for a cloud becomes more expensive and less economical in the long term, just as if you were buying a day ticket every day to commute to work. A monthly subscription would be a wiser choice here.

Cloud computing offers the so-called reserved price model in this situation. This continuous use principle allows you to reserve resources – e.g. 10 virtual CPUs, 1 TB of storage and 20 GB of RAM – for a contractually agreed period of time. With the Open Telekom Cloud, for example, this is possible for 12, 24 or 36 months at a fixed flat rate. Fixed contracts with long-term usage agreements are rewarded with substantial discounts by most providers. In contrast to the pay-per-use model, you still pay full price even if you don’t use all of the resources.

Furthermore, the reserved model can also be divided into two different conventional payment systems. Either the customer pays their subscription for the period of use concluded in fixed monthly installments, or they pay the total amount due immediately and in advance, in which case further discounts can be granted. This makes the reserved model ideally suited to customers who are satisfied with a particular provider and want to commit to using them on a long-term basis in order to benefit not only from the services but also from discounts.

The third pricing model used in the cloud has established itself on the market as a combination of usage-based and fixed costs. This combined model allows cloud users to secure their base load using the reserved model and add additional capacity if necessary, which is then paid for “on demand”, i.e. based on consumption.

This pricing model is an attractive option for companies that only need cloud services on a project-related or temporary basis, are dependent on seasonal business or expect more peaks at the end of the month, for example due to billing runs. In such a case, several servers are added automatically, thus catering to the reality of a business very well.

The National Institute of Standards and Technology, or NIST, has specified five criteria for cloud services, one of which is referred to as measured services. This means that the resource utilization of cloud services has to be measured and monitored and the corresponding resources made available to cloud users. This is the basic requirement for providing users with a transparent list of the services they have used, including traceable costs, on their invoices; otherwise, the provider’s service is not cloud computing according to the NIST definition. 

An analysis of your own requirements is the key criterion when considering the most suitable pricing model. No matter which cloud service and pricing model you ultimately decide on, there is one thing you should always be aware of: As a user of a cloud application, you bear a great deal of responsibility for the costs. With the pay-per-use model, you always ensure that only the resources your company actually needs are used – 24 hours a day, seven days a week.

If you opt for the reserved model, it is important that you pay careful attention to the contract terms and take the notice periods into appropriate account. So choosing the right pricing model alone is not enough – a continuous review of the invoices, some of which can be very complex, is of fundamental importance to ensuring cost efficiency.