Cloud Billing Is Complex and Confusing – Here’s How To Take Control

Companies are used to experiencing sticker shock when they receive their public cloud bills, and it only becomes greater as they dive into the complications of how the bill was produced. Avi Shillo, CEO, Statehub.io, shares his thoughts on why cloud bills are so complicated and what can be done to make them more predictable?

When companies receive the cloud bill from their public cloud provider, they usually can’t tell exactly what they are being charged for. In fact, many companies even put an extra cushion in their budget to pay for unexpected cloud bill prices, with some companies taking into account a margin of up to 30% from what they expect their cloud bill to be.

This leads many DevOps engineers and CIOs to wonder what they can do to better understand their cloud invoice every month.

Cloud Bills Are Cloud Friendly, Not Necessarily Business Friendly

Something important to consider when looking at a cloud bill is that they’re categorized according to public cloud vendor services and not necessarily the service they provide for the business (such as DR and analytics). This is due to the nature of cloud computing; it’s a pay-as-you-go service that simply has too many options “to go”. And since cloud services are services, their consumers (i.e., the engineers) are not required to understand the implementation details of each service in-depth.

All these options lead to too many unpredictable variables to decipher when looking at cloud costs and planning project costs. There are load balancing costs, various monthly costs, cost per GB, various services that are charged by the minute, egress charges, VPN charges, etc. Even services that are sold as “packaged products”, such as managed Kubernetes and database services, end up being billed as a bunch of components that make up the database services.

This makes it a challenge to associate cost with its purpose, understand the return on investment (ROI), or even predict costs. Understanding just one of these items on a cloud bill requires solving complex mathematical equations, which further complicates the process. 

These challenges become exponentially greater in multi-cloud environments.

Learn More: Mission Possible: Operational Flexibility in the Public Cloud

How Did It Get To Be This Way?

The promise of the public cloud is operational expenditure (OpEx), a pay-as-you-go dream that you will only be charged for what you use. In reality, you can actually expand operations without paying in advance (regarding things such as servers, traffic, storage, etc.). But in doing so, you lose the ability to accurately predict how much you will need to pay at the end of the month.

This is how it works across all public cloud providers, as they want to be able to make good on their promise of charging consumers by what they use. 

Therefore, public cloud providers develop complex algorithms that show how much each customer’s usage is affecting their underlying costs, which in turn is reflected in the customer’s complex, detailed invoice as multiple different sections, lines and charges. While this is indeed frustrating from a budgeting perspective, this method of billing frees up engineers’ time to be able to concentrate on other more pressing issues.

Learn More: Reducing the Cost of High Availability in Google Cloud Platform

How To Make Your Cloud Bill More Understandable

Given all this, there are still several ways to make cloud bills more clear and predictable.
From a business standpoint, enterprise tech chiefs can go for the reserved capacity option whenever they can, as it is both cheaper and predictable. Tech chiefs can also work on procedures and checklists within their Ops teams to ensure there are no untagged resources in the system, thereby enabling them to see exactly what each application or workload costs them.

In addition, tech chiefs can look to third-party and managed services that may solve more complex problems, obviating the need for them to assemble the solution themselves from many components and have better predictability in their pricing model. Most managed services consolidate separate cloud services and associated charges and allow customers to easily plan projects according to their goals without focusing on the prices of each component of their deployment. This further simplifies predictability of cost (for example, Statehub for multi-cloud and cross-region business continuity).

Finally, tech chiefs should strive to keep themselves vendor-independent. They should choose portable technologies and open standards that they can deploy on multiple clouds whenever possible. For example, Kubernetes makes it simple to deploy and maintain databases, making the advantages of a managed database significantly less relevant. This means they can more effectively predict their costs by calculating the basic resources (VMs, storage, throughput) they need.

Due to all the moving parts of a cloud bill, it is unlikely its structure will change anytime soon. However, by using managed services and taking advantage of various public cloud options, companies can begin to have a better grasp of how much their bill will be at the end of every month and fully take advantage of public cloud services.

Did you find this article helpful? Tell us what you think on LinkedInOpens a new window , TwitterOpens a new window , or FacebookOpens a new window . We’d be thrilled to hear from you.