AWS Smart Business Blog

Four Principles of Cloud Financial Management Small and Medium Business Owners Need to Know

IT can be a daunting topic—especially if you are a small or medium business (SMB) leader and do not consider yourself a technical person. Traditionally, IT costs are centered around up-front capital investments on hardware and infrastructure, and assembling a team aligned to support and maintain that infrastructure. There are also energy and premises costs to consider to host all this infrastructure. The business owners and technology teams could submit requests for certain hardware and services, but essentially this model relies on organizations having the capital to invest upfront, and leaves the business constrained by buying decisions made before solutions were fully understood. As you can tell, it’s a lot to contend with.

A fundamental benefit of using cloud for your business operations is not having to invest up-front; infrastructure and services are available on-demand on a pay-as-you-go basis. Even when using cloud-based software, a central set of services is often still required to knit all these together. This might include connecting customer relationship management (CRM) data to website sales data for example or even invoicing systems. Regardless of the type of cloud usage an organization has, they cannot build without any regard to cost forecasting, monitoring, and optimization. In fact, continuous cost optimization is essential to maintain profitability.

Take for example Runtastic, an SMB which migrated to Amazon Web Services and saved €300,000 in new infrastructure cost alone. They protected their growth and delivered rapid scaling when they were acquired by Adidas.

Our goal is to provide SMBs with a mental model to de-mystify cost optimization and prepare to set cost targets. These conversations can be valuable for internal IT colleagues or external tech partners.

An easy way to understand fixed and variable IT costs

To help you get started, let’s think about grocery shopping. Even if you aren’t the one going to the store each week or paying the bill, you are likely to know how much your grocery budget is and how much your household spends each week. You’ll likely also know who’s consuming what from the groceries, and how you individually contribute to the overall consumption and cost. This is a fixed cost.

There are a few common factors that affect how we shop. There will be consistent, go-to ingredients you buy each week, but there are always some fluctuating situations, like visitors coming for dinner, a change in meal rotation as the seasons move from autumn to winter, or purchasing a seasonal holiday spread. In essence, your choices are based on your staples, plus a host of variables such as time of year, budget, and seasonal availability. Importantly, the budget and purchasing decisions are always transparent, and adjustments are made as and when they’re needed, based on demand. This is a variable cost.

So, how does this translate to cost management for cloud computing?

Defining cloud financial management

Cloud financial management (CFM) is a newer discipline and way of working that helps SMBs to:

  • Allocate and track cost and data usage
  • Simplify IT costs with consolidated billing
  • Enable better planning, billing, budgeting, and forecasting
  • Optimize cost by reallocating services to better serve the business

CFM is sometimes referred to as FinOps, which is short for financial operations. This is terminology your technical counterparts might already know. In the following video, Mike Blackmer of AWS Training and Certification further explains the concept for SMBs such as yourself.


A metaphor to understand the guiding principles of cloud cost management for SMBs

Let’s take a closer look at some elements that drive our grocery shopping patterns, and how they easily transfer to CFM (or FinOps) in your business.

1. Staples

We all have staples we buy regularly—such as bread and milk—and have an idea of what they cost. In the cloud world, this will most likely be the services that run your most core software and it won’t change often such as your sales website, data archives, and contact center software. You’re unlikely to turn your website off any time soon, so it’s fairly straight forward to establish a long term run rate and cost estimate. Exactly which AWS services are used will vary from company to company. In any case, these foundational services are always there at a baseline level.

We advise to start with optimizing these basics, understanding what your baseline cost is, and setting out to minimize that. For example, by making sure you do not have more capacity than you need (right sizing), and then making longer term commitments to the capacity to get a better deal. On AWS, that could be applying a savings plans or purchasing reserved instances. Either your AWS Partner Network consultant, external tech vendor, or engineering teams could help you with this.

While planning a two-year region migration, Upwork had the certainty needed to make a long term commitment through savings plans, and saved 45% on compute costs during the migration and beyond

2. Frugality

Everyone likes to be as economical as possible when grocery shopping, such as buying in bulk, taking advantage of store loyalty points, or choosing the most value for money products. The same should apply in the cloud.

There are discounts for making longer term commitments as we saw with the staples above. AWS offers price advantages that will apply depending on your particular use case. For example, if you have a dataset you need to keep for compliance, but don’t access often—such as audit records—you can pay less to store it by committing to the less frequent access. For example, CineSend has an average cost savings of 33% per month by using our storage service, called Amazon Simple Storage Solution (S3). Within S3, they use our automatic intelligent tiering to select the correct storage type for their data based on usage.

3. Variability

We all have to cope with variation in demand. During the holiday season we have to make sure there’s enough food to go around when hosting friends and family. Inversely, you don’t buy or use groceries when you go away for any period of time on vacation. The ability to only pay for what you use is one of the core benefits of adopting AWS Cloud. If you have a seasonal business, you can choose to scale capacity up and down based on known demands. Additionally, features such as auto-scaling allow your capacity to scale for unexpected demand, too.

To embrace variability, start by understanding how much of your capacity is being used right now, and then identify where you expect highs and lows in traffic through the year. How will you scale and up and down for those events? How frequent the peaks are will help you answer this question. It’s not very frugal for example to have IT vendors or in-house staff scaling down every Friday, and back up again on Mondays – that could be automated.

Here’s a good example from Echelon Fitness, an SMB which was able to automatically add or remove compute power depending on traffic to meet 1,000% annual growth during COVID-19 demands.

4. Experimentation

Sometimes we like to try something new, or have a specific short-term requirement. Perhaps a household member wants to switch to more protein or fewer carbs, or they want to try new breakfast cereal. Either way, the alternatives and budget will need to be considered. In the cloud world, this is welcome and happens all the time as organizations experiment and innovate, or new services and instance types are launched.

For example, a business unit might want to experiment with new ways to forecast demand for raw materials for a manufacturing process. Rather than having to invest upfront in cloud-based software, or their own hardware and software (and the skills needed), the services needed can be spun up quickly in the cloud, often with help from the AWS community. If the experiment doesn’t live up to the hypothesis, you can quickly shut down and look to try something new another time. If the experiment is a success, you can look to build the new services into your staples and apply the frugality principles.

Experimentation drives innovation culture within a business and helps to create differentiating features. In a traditional model, the cost of entry can be a barrier to experimentation. You may need to procure hardware, have this installed and configured by your IT team and, if the experiment fails, you have potentially invested a large amount of capital that could end up wasted. If the experiment is successful, you may only have bought enough resources to cover the experiment with and so you’ll need a second round of procurement to scale up, delaying your speed to market. With the cloud model, you can spin up small compute instances at low cost to experiment with and, if the experiment is a success, you can scale up quickly and painlessly, or if the experiment did not work, you can simply switch off the resources and pay nothing further.

New to cloud? Create an AWS Free Tier account and take this 10-minute beginner-level course to learn how to control your cloud costs.

Banner driving to our interactive assessment tool

Key takeaways to starting your cloud cost optimization journey

Now that you’ve had an opportunity to explore some of the principles that drive cost management, it’s time to learn about the four key pillars of CFM. A structured approach to FinOps will support your business transformation through forecasting, cost transparency, control, and optimization.


Understand your cloud setup and who owns all the services. You need to have the ability to plan and set expectations around cloud cost for your projects, applications, and more. AWS provides forecasts based on your cost and usage history and allows you to set budget threshold and alerts. Remember: measurement and accountability within your teams is key.

Track and allocate

You need complete, near real-time visibility of your cost and usage information to make informed decisions. With AWS, you can provide real-time cost data that makes sense to your engineering, application, and business teams, establishing a culture of cost transparency and ownership. Showing where certain business areas spend—such as website hosting and contact center software—is much more impactful than displaying the overall cost without context.


SMBs are always challenging themselves to reduce spend and optimize overall value. It’s a healthy conflict that can drive the most balanced outcome. It should be no surprise that the same goes for cloud cost optimization. It’s an ongoing conversation that SMBs shouldn’t be afraid to have. Instead, we at AWS lean into it.


Learn more on how to plan and evaluate your cloud spend. Use forecasting to manage and control your costs. With agility throughout the cloud, gaps in data governance can occur. It is possible that an individual can accidentally increase the bill by leaving resources on when they should have been shut off. You need not only need a simple way to access and visualize comprehensive billing information, but appropriate guardrails must be set so you can establish control over cost, governance, and security.


You can now run your FinOps practice with confidence.

Next steps

We hope the grocery metaphor is helpful in making cost management in the cloud more approachable, and in inspiring you to start a structured approach. We have another blog post if you’re more interested in basic education on how SMBs can better manage their cloud costs.

Ready to dive deeper into the world of CFM? Take a closer look into SMB-specific training, AWS Cost Management , or contact us to speak with an SMB expert.

Mathew Godfrey

Mathew Godfrey

Mathew Godfrey is a Sr. Customer Solutions Manager at AWS, guiding Enterprise customers in UKI on their cloud adoption journeys. With almost two decades of IT delivery and transformation experience, Mathew works with customers to bring best practice to large scale migrations. He helps them build their organizational cloud maturity and to apply cost optimization best practices. Mathew is based in the UK.

Dave Walters

Dave Walters

Dave Walters is a Senior Solutions Architect at AWS, focusing on the retail industry. Dave has over 20 years of experience working in IT infrastructure. He has worked for several well-known online brands, leading teams through the transition from hardware, through virtualization, and to cloud.

Sandrine Palmer

Sandrine Palmer

Sandrine Palmer is a Client Partner within AWS Professional Services focusing on enabling customers to accelerate and realize the value of the cloud. Sandrine previously worked as a Senior Business Development Manager in the AWS EMEA Cloud Economics practice where she focused on Cloud Financial Management.