Amazon Web Services (AWS) Pricing Model and Plan Guide, in the UK

With its numerous pricing models available, many UK businesses struggle to navigate the complex AWS pricing. From pay-as-you-go plans to savings schemes, AWS offers a bit of everything, just not in the clearest way. And if you're paying in foreign currencies, you should be wary of any extra fees. A Wise Business multi-currency account can help you dodge unwanted fees while keeping things simple.

Get started with Wise Business 🚀

AWS models pricing overview

AWS (short for Amazon Web Services) is Amazon's cloud computing platform¹. It powers everything from startup websites to enterprise apps and massive machine learning models. Instead of buying servers or building your own data centre, you rent what you need – compute power, storage, databases, and more, and it's on demand.

As for the AWS pricing plans, they are flexible, and you only pay for what you use. That's the idea, anyway. The reality is that there's a lot to choose from, which is where things get tricky (and pricey).

In practice, storage, compute, and outbound data transfer are the main cost drivers on an AWS bill². Inbound data transfers and inter-service transfers in the same region are generally free. For example, AWS notes that EC2 and RDS (compute/database) usage is charged by the hour or second, while S3 storage and data egress are billed per GB.

How do AWS pricing and models work?

So what does it mean to get billed based on what you actually use? It could be based on the actual resources you consume, such as the seconds of EC2 compute, GB‑months of S3 storage, or Lambda invocations by the millisecond. The pricing differs by service type – IaaS services like EC2 and RDS are metered by time, storage services like S3 charge per GB‑month, while PaaS/serverless offerings like AWS Lambda bill per request and compute time.

For instance, EC2 billing starts in one‑second increments (with a 60‑second minimum) for on‑demand use. AWS also offers Savings Plans and Reserved Instances to reduce costs when you commit to a certain usage over 1 or 3 years. These can cut compute costs by up to 72% vs on‑demand³. Additionally, AWS lets you save by using more, and higher volumes of storage or data transfer bring tiered pricing discounts.

AWS pricing principles

So here's the thing – there are no upfront fees or long-term contracts unless you choose them. The platform highlights three core pricing principles: "Pay-as-you-go," "Save when you commit," and "Pay less by using more." These principles give businesses the flexibility to scale and save based on how much they use and how long they plan to stick around.

"Pay as you go" pricing

This is AWS at its most flexible. With pay-as-you-go pricing, you only pay for what you use, without any contracts or commitments. Want to try an EC2 instance for an hour? You'll pay for an hour (or even just a few seconds, depending on the instance). Or store 100 GB in S3 for a month? You'll only pay for that storage, down to the byte. It's ideal for businesses that want to test ideas, run short-term projects, or need to scale up and down often.

"Pay as you go" benefits

This model is great for businesses that value flexibility over long-term commitment⁴. You're free to scale up, scale down, or walk away whenever you need to because there are no strings attached. Some important benefits include:

• No upfront costs, and you only pay once you start using a service
• Perfect for testing, pilots, or seasonal workloads
• Real-time cost control, as you can stop resources instantly and avoid overspending
• Most services are billed per second, so you don't pay for unused time
• Ideal for startups and small teams that need room to grow

"Save as you commit" pricing

If you know you'll be using AWS regularly, this model rewards you for committing⁵. Instead of paying the full on-demand rate, you agree to use a certain amount of services (measured in $ per hour) over 1 or 3 years, and in return, you get a hefty discount. It's a way to trim costs without giving up too much flexibility, especially with Savings Plans, which apply across multiple services and instance types.

"Save as you commit" benefits

This pricing model can be a smart move if you're ready to plan ahead. This model works well for steady, predictable workloads like production environments or always-on apps. Here's why businesses often go for it:

• Lower long-term costs⁶, where you can save up to 72% compared to on-demand pricing
• Flexible plans, where you get to choose 1- or 3-year terms, with options for upfront, partial upfront, or no upfront payment
• Applies across services, especially with Savings Plans, and discounts can cover EC2, Fargate, Lambda, and more
• Ideal for apps or services that run consistently

How to pay for AWS, 6 Pricing Models

A pricing structure is basically how AWS lets you pay for what you use. There's no flat monthly fee like with many traditional SaaS providers. Instead, you choose a model that fits how your business runs. Below are the six AWS pricing models, with examples, pros, and cons to help you figure out which one works best for your needs⁴.

Free Tier pricing

The AWS Free Tier is Amazon's way of letting businesses try out its services without committing to immediate costs. It offers limited access to a wide range of AWS products, so it's perfect for early-stage startups, side projects, or teams that want to test features before scaling.

The Free Tier is split into three types: Always Free, 12 Months Free, and Trials. For example, new users get Amazon EC2 t2.micro or t3.micro instances for 12 months – 750 hours/month. They also get Amazon S3 storage (5 GB) and 1 million requests with AWS Lambda⁴.

As for AWS API pricing, keep in mind that the free version of Amazon API Gateway comes with 1 million API calls (that's for REST APIs), and more.

Pros of Free Tier

• Great for experimenting without spending money
• Ideal for MVPs, student projects, or learning
• Access to popular services like EC2, S3, RDS, and Lambda

Cons of Free Tier

• Usage limits are tight (for example, 5 GB S3 isn't much for media-heavy apps)
• Easy to accidentally go over limits and incur charges
• Not suitable for production environments in the long term

A UK-based micro-business could build and test a simple website or app backend using Free Tier services, without the need to touch their budget.

On-demand pricing

On-demand pricing is AWS's most flexible and simplest option. You spin up resources (like virtual machines or databases), use them for as long as you need, and only pay for what you consume. You pay for the second, minute, or hour, depending on the service. There are no upfront costs, long-term contracts, or hidden fees.

For example, if you launch an Amazon EC2 instance to run a few data analyses and shut it down after 6 hours, you'll only be billed for those 6 hours. Similarly, with services like AWS Lambda, you're charged based on the number of requests and execution time⁴.

Pros of On-demand

• No commitments or upfront payments
• Great for short-term projects or unpredictable workloads
• Easy to start and stop resources as needed

Cons of On-demand

• Higher cost per unit compared to reserved or committed options
• Not ideal for stable, consistent workloads

Let's say a UK-based e-commerce company needs extra computing power during Black Friday. On-demand lets them scale up temporarily without being locked into a year-long contract.

Spot instance pricing

Spot instances will allow you to use AWS's unused cloud capacity at significantly reduced rates, up to 90% more affordable than on-demand pricing. The catch is that these resources can be reclaimed by AWS at any time with little notice. You bid on the compute capacity, and as long as your bid exceeds the current spot price, your workload runs.

Billing is per second or per hour (depending on the service), and once your instance is interrupted or terminated, charges stop immediately. As such, it's best for flexible applications⁴.

Pros of Spot instance

• Large cost savings, ideal for budget-sensitive projects
• Perfect for batch jobs, data analysis, CI/CD pipelines, and machine learning training
• Works well when combined with automation or backup strategies

Cons of Spot instance

• Instances can be interrupted at any time
• Not ideal for critical, long-running workloads

A startup training a machine learning model on a tight budget can use Spot instances to run compute-heavy workloads overnight. If interrupted, the job can be resumed later or restarted.

Reserved instance pricing

Reserved Instances (RIs) are a way to lock in AWS compute capacity for 1 or 3 years in exchange for a significant discount that goes up to 72% off compared to on-demand prices. You commit to a specific instance type, region, and operating system, and in return, you get predictable costs and guaranteed availability (especially helpful in high-demand regions).

Billing is upfront, partial upfront, or monthly over the term. Once purchased, RIs can't be cancelled, though AWS does offer a marketplace to sell unused reservations.

Pros of Reserved instance

• Big cost savings for steady-state or long-term workloads
• Predictable pricing and easier budget planning
• Capacity assurance in specific regions

Cons of Reserved instance

• Requires long-term commitment
• Less flexibility because it's locked to instance specs and region
• Potential for waste if usage changes

A SaaS business running the same backend infrastructure 24/7 can save money by reserving compute instances for the next three years.

Savings Plans pricing

AWS Savings Plans offer a flexible way to save up to 72% compared to on-demand pricing⁷, just like Reserved Instances, but without locking into specific instance types. Instead, you commit to a consistent amount of usage (measured in $/hour) over a 1- or 3-year term, and AWS applies discounted rates automatically across matching services. There are two types:

• Compute Savings Plans (most flexible – apply across EC2, Lambda, Fargate)
• EC2 Instance Savings Plans (less flexible – tied to instance family, region, OS)

Pros of Savings Plans

• Significant savings over on-demand
• More flexibility than Reserved Instances
• Supports a broader set of services

Cons of Savings Plans

• Still requires time-based commitment
• Underusage, meaning wasted spend
• Slightly more complex to manage

A business running multiple services (like EC2, Fargate, and Lambda) across various regions can use a Compute Savings Plan to enjoy discounted pricing without needing to guess future instance types.

Dedicated Hosts pricing

AWS Dedicated Hosts pricing is designed for businesses that need physical servers dedicated entirely to their use. They often need it to meet compliance, licensing, or security requirements. Rather than sharing infrastructure with other customers, you rent a full EC2 server⁸. Billing is based on the host (not instance), and you can choose on-demand, reservation, or Savings Plan payment options.

AWS bills you per host, per hour (or per second, depending on the OS), regardless of how many instances you run on that host. This can be cost-effective if you maximise use.

Pros of Dedicated Hosts

• Great for compliance-heavy industries (finance, healthcare)
• BYOL (Bring Your Own License) support for Windows Server, SQL Server
• Full control over instance placement

Cons of Dedicated Hosts

• More expensive than shared EC2 in low-usage scenarios
• Higher complexity in capacity planning
• Not always necessary for standard workloads

A financial services company with strict regulatory requirements might use Dedicated Hosts to ensure workloads run in physically isolated environments.

How to save on your AWS bill next month

Running AWS can be cost-effective, but only if you stay on top of it. For UK businesses, AWS pricing is flexible but complex, so it pays to be in control. You can start by choosing the right pricing model for your business needs. From there, regularly audit your usage with your team.

One more tip is that you should watch out for hidden fees. If you're being billed in USD or another foreign currency, foreign transaction charges from your bank can sneak onto your monthly AWS invoice. Avoid them by paying with a Wise Business multi-currency card, available to pay in 40+ currencies. It lets you pay for platforms like AWS easily with low fees and no exchange rate markups, always at the mid-market rate. Get a Wise Business account to save up smartly.

💡 Read more on Amazon Web Services billing

Manage AWS payments with Wise Business

AWS pricing can seem complex at first, but understanding its flexible models helps you choose what fits your needs and budget best. Whether you're just starting out with the Free Tier or committing to Savings Plans for long-term savings, there's an option for every UK business.

But most importantly, keep an eye on your usage, pick the right plan, and don't forget that hidden fees, like currency conversion, can add up. That's why a Wise Business multi-currency card is a smart way to pay your AWS bills with low fees and the mid-market exchange rate*, so you can save even more.

Register with Wise Business🚀

FAQ

What is the pricing currency?

AWS pricing is listed in USD by default, though some regions and billing accounts may offer local currency options. UK businesses often pay in USD, so foreign exchange fees can apply.

What is the default pricing model?

The default model is pay-as-you-go, where you pay only for the resources you use, with no upfront fees or contracts.

Does AWS have any implementation and onboarding costs?

AWS itself doesn't charge for onboarding or setup, but some complex enterprise services or partner solutions might. Businesses should factor in possible consulting, migration, or managed service fees when budgeting overall costs.

---

Sources used:

1. AWS Product and Service Pricing – Amazon Web Services
2. Key Principles - How AWS Pricing Works – AWS Documentation
3. AWS Pricing: 5 Models & Pricing for 10 Popular AWS Services – Spot.io
4. Choosing The Best AWS Pricing Model: A Complete Look At AWS Pricing (With Examples) – CloudZero
5. AWS Product and Service Pricing – Amazon Web Services
6. Compute and EC2 Instance Saving Plans – Amazon Web Services
7. Cloud Cost Savings - Savings Plans – Amazon Web Services
8. Amazon EC2 Dedicated Hosts – Amazon Web Services

Sources last checked: 07-Aug-2025