Your smart guide to AWS cost optimization

More than the tools you use to monitor your cloud spend, the key to sound cost management lies in the relationship between you and your trusted AWS partner.

The competitive benefits of migrating and managing workloads on the cloud are overwhelming, and the AI revolution is raising the stakes.  

But costs are rising quickly. Global cloud expenditure is expected to jump from just over $750 billion in 2024 to $2.39 trillion by the end of the decade – and business leaders are increasingly anxious that their control over these costs is imperfect. 

Almost two-thirds of FinOps executives are unable to report accurately on key metrics of cloud spending, while engineers are seeing greater levels of disruption to their work as a result of poor cloud cost management. 

In other words, optimization of your cloud spend is a vital FinOps function.

The world’s largest cloud provider, Amazon Web Services (AWS), gives businesses the tools – and often the financial backing – to improve visibility into their cloud usage and spending. This gets you the insight to redress any imbalance between the two and find the sweet spot for your use case among a wide choice of AWS pricing models. 

In this blog post, we show you the best ways to achieve effective AWS cost management – not only as you migrate but as your needs and spending profile change over time. 

AWS cost optimization tools are free of charge with your cloud contract, but it is up to you to implement and manage them, which can rapidly become daunting. Most organizations rely on a partner or partners to control AWS costs either as they migrate or at a later stage when, as existing AWS users, they perceive a need to curb costs. 

We start our briefing on AWS cost optimization by discussing its main pricing models and the pros and cons of each as you apply them to your use case and expected needs. A skilled partner will help you land on a pricing model that is advantageous to you and then help you determine your actual usage and the extent to which you are overspending, or could be better off on a different pricing model. To do this, you need to continuously monitor your AWS costs and usage as you build your relationship with AWS and your partner. 

You expect your partner to understand the available AWS tools, and to implement them intuitively – but more than that, you want them to fight your corner with AWS. 

The strength of SoftwareOne (formerly Crayon) is that it can (re)architect your AWS presence, build cloud resilience, advise on compliance, arrange cloud storage, take care of AWS cost management as part of its ongoing support – and help you secure deeper discounts because of its close and formal ties with AWS. 

AWS cost optimization is never a one-off but a business process that should be part of your everyday FinOps practices. To build and refine these habits of cost management and make them second nature requires trust. The key to AWS cost optimization is not the successful outcome of a project but the strength of the relationship between yourself, AWS, and your trusted AWS partner. 

Migration – or how to optimize as you mean to go on

Migration is a highly technical process, so you require your AWS partner to be capable of building your cloud and helping you find the right pricing model because both are essential. 

We shall take it as read that a poorly designed migration will have negative ramifications on performance and costs and focus on the choice of AWS contract. 

The graphic below sets out the standard AWS pricing models and their respective advantages and disadvantages. 

AWS Pricing Model Analysis

Pricing Model	Pros	Cons
On-Demand	No long-term commitment, flexible	More expensive than reserved pricing.
Reserved Instances	Significant savings (up to 72%).	Less flexibility
Savings Plans	Flexible savings across EC2, Lambda, and Fargate.	Commitment required (1 or 3 years).
Spot Instances	Extremely low cost (up to 90% cheaper).	Can be terminated by AWS at any time.

On-Demand is often a great “starter model” for businesses seeking to understand their cloud needs without incurring upfront costs. Once they do, it is straightforward to switch to another pricing model because the On-Demand option is highly flexible. 

Whereas On-Demand lends itself well to applications that have unpredictable spikes in load, the Reserved Instances pricing model is suitable for running legacy applications that are predictable in their usage, which can be planned well in advance. 

In this model, you can reserve instances for a period of one to three years at a discount of up to 72%. However, you are committed to these instances and pay for them even when you scale down.

In the Spot Instances pricing model, the situation is reversed: you bid for spare capacity on Amazon’s open market, leading to deep discounts of up to 90%. But if the capacity is not available, the Amazon EC2 Spot service can interrupt your instance, making this model suitable for businesses with advanced cloud-native development capabilities. 

Amazon distinguishes various types of Savings Plans. Of these models, Compute Savings Plans are the most flexible as they can get you savings of up to 66% across EC2, Lambda, and Fargate – a serverless, pay-as-you-go computing engine. The attraction of this approach is that you benefit from discounted prices regardless of instance size, region, instance family, operating system, or tenancy.

EC2 Instance Savings Plans offer deeper savings of up to 72% in exchange for commitment to usage of individual instance families in a region. This automatically reduces your cost on the selected instance family in that region regardless of Availability Zone, size, operating system, or tenancy.

Businesses can also negotiate bespoke contracts, so-called AWS Private Pricing Agreements (PPAs), which we shall discuss in a separate section. 

Your choice of contract is not set in stone. A huge aspect of AWS cost optimization is to monitor usage and potentially adjust or switch pricing models in light of your findings. The sooner you start embedding cloud control processes in your FinOps operations, the earlier you will be able to rightsize cost and consumption. 

Finally, one technical aspect of migration with a bearing on AWS cost management is the development of a landing zone. The goal of a landing zone is to provide AWS users with a starting point for application or workload deployment. Landing zone accelerators allow businesses to see how costs are distributed across accounts—a hugely relevant insight that two-thirds of businesses do not currently have. 

Why do cloud costs overshoot, and how do you mitigate against it?

The imbalance between expected and actual patterns of cloud consumption is the number one target for AWS cost optimization. Businesses often over-provision. They are running instances with low CPU utilization, have contracted for larger instances than necessary, pay for storage that is unused or have overpaid for high-performance storage when lower-cost alternatives such as S3 Glacier would have done the job just as well. 

Other mismatches are caused by misconfigured Reserved Instances failing to align with actual usage patterns or a lack of auto-scaling to rightsize resource allocation with the level of demand. 

While you should not consider cloud optimization as a single project with a single great outcome, SoftwareOne always finds significant immediate savings for AWS customers embarking on their cost management journey. For a $1 billion hotel group, this saving was of the order of 20%, and an international insurance company reported a one-off reduction of $200,000 in their AWS costs.

These economies are great news; they are the happy start to your AWS optimization journey.

AWS has a barrage of optimization tools that redress the balance between projected and actual use, each with its extensive documentation. In and of themselves, these solutions are fairly straightforward, but there is complexity in the sheer number of them and in the challenge of orchestrating them effectively to gain meaningful insight to start bringing down costs. 

To give you a flavor of some of that complexity, we list the most commonly used AWS optimization solutions below: 

• AWS Cost Explorer: Analyzes past usage and forecasts future costs.
• AWS Budgets: Sets and enforces cost thresholds.
• AWS Trusted Advisor: Identifies underutilized resources and security gaps.
• AWS Compute Optimizer: Provides right-sizing recommendations.
• AWS Savings Plans and Reserved Instance Reporting: Helps manage cost-saving commitments.

Organizing this constellation of tools (and there are many others) is impractical for most businesses, and engaging with a recognized AWS optimization partner is an important milestone of cloud maturity. 

With more than 350 AWS Certifications across nine AWS Competencies, as well as three Service Delivery Validations, SoftwareOne is one of the most authoritative players in the space. It has integrated the FinOps aspects of the AWS toolkit on a dedicated cloud optimization platform, Cloud Cost Control. 

This multi-cloud solution is used by SoftwareOne’s AWS team to generate real-time transparency into AWS cloud costs and to present businesses with the data and visibility they need to control their spend. 

SoftwareOne’s reach in AWS optimization projects goes beyond its technological edge; it can also leverage its relationship with AWS to secure funding for businesses that want to put their house in order by establishing cogent and well-architected cost management practices. 

This ability to free up funding is an example of SoftwareOne’s soft power, which extends across many areas of AWS optimization and cost management. One such area is AWS Private Pricing Agreements (PPAs).

The power of PPAs

By offering a wide range of pricing models, AWS gives you the best chance of optimizing cloud costs for your use case, and if your business satisfies certain conditions, you can even negotiate a bespoke deal. To qualify for a PPA, you must: 

1. Spend over $500,000 on AWS services annually;
2. Commit to the PPA for at least a year (though a fixed term of three or five years will likely secure a deeper discount);
3. Expect to grow cloud usage by typically 20% year-on-year.

For most larger businesses, the latter condition is not an insurmountable hurdle, given that the global cloud computing market is forecast to post a CAGR of 20.4% between 2025 and the end of the decade. 

You can negotiate pricing with AWS on your own, but you will be missing out if you go down this road. With SoftwareOne in your corner, you can achieve optimal terms through its unrivaled knowledge of the ins and outs of AWS pricing structures. The soft power of SoftwareOne as a Premier Tier AWS Partner will often yield additional discounts while giving you strategic guidance and ongoing support in your PPA journey. 

The optimization benefits of a PPA do not extend to cloud spend alone; you can use the required $500,000+ annual spend to make third-party software purchases on the AWS Marketplace. 

You can do this through Amazon or, better still, through SoftwareOne’s Cloud-iQ platform. Instead of purchasing the software directly at list pricing, you can request a private offer from SoftwareOne, unlocking exclusive savings. 

In this way, a cornerstone of AWS cost optimization flows naturally to other cost benefits, brokered by SoftwareOne’s relationship with AWS. 

It doesn’t stop there; Cloud-iQ is instrumental in another crucial aspect of FinOps optimization – managed billing. 

Managed billing

Before you can start optimizing, FinOps must understand the costs of cloud consumption – and your bill will tell you where your money is going. 

AWS submits its bill to you in exacting detail. Its invoices, especially Cost and Usage Reports (CURs), can be extremely large, often containing millions of rows because AWS charges for consumption by the second. Detailed billing unlocks insight into usage and spending patterns, but the sheer amount of detail can be overwhelming and sometimes too extensive for even spreadsheets to handle. 

SoftwareOne’s managed billing makes this process much simpler and more intuitive. You receive a single, consolidated monthly billing statement that captures all your AWS expenses, giving FinOps a clear view of cloud costs across the organization. 

Through SoftwareOne’s Cloud-iQ portal, you can access detailed billing and reporting tools that make it easier to understand and manage your costs. Our AWS Cost Control Dashboard gives you real-time insights and recommends cost-saving adjustments based on actual usage. This means you can make informed decisions without wading through endless billing data. 

Transferring the billing for your AWS accounts is typically no more complicated than a handover of your root user email to SoftwareOne’s email. This is all it takes for AWS to recognize that the consumption is to go through SoftwareOne instead of directly to you, and from that moment, the discount agreed between AWS and SoftwareOne is passed on to you. While it is hard to isolate the effect of this discount in the ever-changing pattern of your AWS consumption, businesses are seeing significant savings of between 5% and 10% as the outcome of their managed billing arrangement with SoftwareOne.