If you do not understand how AWS charges you, you could end up with a £10,000 bill for services you thought were free. Cloud computing is not magic money — every virtual server, every hour of database use, and every gigabyte of storage has a price. This chapter explains the different ways AWS lets you pay (so you can choose the cheapest option) and exactly what the Free Tier offers beginners like you who want to experiment without fear.
Jump to a section
You walk into a restaurant with £20 in your pocket. The restaurant offers two ways to eat: an all-you-can-eat buffet for £15, or an à la carte menu where you pay £5 for a starter, £8 for a main course, and £3 for a drink. If you are very hungry and plan to eat three courses, the buffet saves you money. But if you only want a single main course, the à la carte option costs just £8 — less than the buffet. Some restaurants also offer a free taster plate: one small starter, completely free, every single day. That free taster is their way of getting you in the door so you might come back and pay later.
This is exactly how AWS Pricing Models and Free Tier work. AWS gives you a choice between paying only for what you use (à la carte, called on-demand pricing) or committing to a longer-term contract for a discount (the buffet, called reserved pricing). The Free Tier is the free taster plate — a limited amount of AWS services you can use without paying, for a limited time or a limited quantity each month. The trick is that once the free taster runs out, you start paying full price unless you plan carefully.
When you study for the CLF-C02 exam, you need to understand that choosing the wrong pricing model can burn through your budget very quickly. A business that runs a server 24/7 but pays on-demand rates is like eating the à la carte menu every single day — it adds up fast. The exam expects you to know which pricing model fits which situation, and exactly what the Free Tier includes and what it does not.
AWS Pricing Models are the different methods that Amazon Web Services uses to charge customers for using its cloud services. Think of them as different payment plans for renting computing power, storage, databases, and other IT resources over the internet. Before cloud computing existed, companies had to buy physical servers, which cost thousands of pounds upfront and took weeks to arrive. AWS changed this by letting you rent exactly what you need, for exactly as long as you need it, and only pay for what you actually use. This is called a pay-as-you-go model — you pay by the hour or by the second, with no upfront costs or long-term contracts.
There are three main pricing models you need to know for the CLF-C02 exam: on-demand, reserved, and spot instances. Each model is designed for different usage patterns and budgets.
On-demand pricing means you pay for compute or storage capacity by the hour or by the second with no long-term commitment. You launch a virtual server (called an Amazon EC2 instance), use it for a few hours, and then terminate it — you only pay for those hours. This is ideal for workloads that are unpredictable, like a new application that might get popular overnight or a short-term data analysis job. The downside is that on-demand is the most expensive way to run something continuously, because you are paying the full list price every hour.
Reserved pricing is like buying a season ticket for a train rather than paying for a single ticket every day. You commit to using a specific AWS service (like an EC2 instance type or an RDS database) for a one-year or three-year term, and in exchange, AWS gives you a significant discount — up to 72% off the on-demand price. There are three payment options within reserved pricing: all upfront (pay everything now for the deepest discount), partial upfront (pay some now, some monthly), and no upfront (pay nothing now but still get a decent discount). Reserved pricing is best for predictable, steady workloads like a company's main website database that needs to run 24 hours a day, 7 days a week.
Spot instances are the most dramatic cost-saving option. AWS has spare computing capacity that nobody else is using at that moment, and it auctions this capacity off to customers at huge discounts (up to 90% off on-demand). The catch is that AWS can reclaim that capacity with only two minutes' notice, so your application might be interrupted. Spot instances are perfect for fault-tolerant workloads like batch processing, data analysis, or rendering video frames — jobs that can be paused and resumed without breaking anything.
Another important pricing component is data transfer costs. AWS charges for data moving out of its cloud to the internet (egress), but data moving into AWS from the internet (ingress) is usually free. Data transfer between AWS services in the same region is often free or very cheap, but transferring data between regions costs money. This is a common exam trap — students assume all data movement is free, but it is not.
Now let us talk about the AWS Free Tier. The Free Tier is a limited, free usage allowance for many AWS services. It is designed to let new customers (and hobbyists) try AWS services without spending money. There are three types of Free Tier offers: always free, 12 months free, and short-term trials.
Always free offers never expire. For example, AWS Lambda gives you 1 million free requests per month forever. Amazon DynamoDB gives you 25 GB of free storage permanently. These are great for small personal projects or learning.
12 months free offers start when you create your AWS account and last for one year. The most famous example is the t2.micro or t3.micro EC2 instance — you get 750 hours per month of that specific instance type for free. That is enough to run one virtual server continuously for the entire month. After 12 months, you start paying full price.
Short-term trials last from a few weeks to a few months. For example, Amazon SageMaker (a machine learning service) gives you 250 hours of free usage for the first two months. These trials are meant to let you test a service before deciding to pay for it.
The most important thing to understand is that the Free Tier is not unlimited free usage. If you exceed the free allowance (like running a bigger EC2 instance than t2.micro, or running two instances instead of one), you will be charged. Many beginners accidentally incur bills because they assume everything is free. The exam tests whether you know the exact limits — for example, that the EC2 Free Tier only covers one specific instance type (t2.micro or t3.micro), not all instances.
Finally, AWS also offers Savings Plans, which are like reserved pricing but more flexible. A Compute Savings Plan lets you commit to spending a certain amount per hour on compute services (like EC2, Lambda, and Fargate) for one or three years, and you get discounts in return. The difference from reserved pricing is that Savings Plans automatically apply to any instance type or region, giving you more flexibility.
To summarise: AWS pricing models give you choices based on how predictable your workload is. On-demand is flexible but expensive. Reserved pricing and Savings Plans save money on steady workloads. Spot instances are cheapest but interruptible. The Free Tier lets you learn and experiment for free within strict limits. As a CLF-C02 candidate, you need to match the right pricing model to the right business scenario and understand exactly what the Free Tier covers so you do not get an unexpected bill.
Choose a Pricing Model Based on Workload Predictability
Analyse your application: if it runs 24/7 and is predictable (like a company website database), choose reserved or Savings Plans for the best discount. If usage is unpredictable (like a new app with unknown traffic), choose on-demand for flexibility. If the job can be paused (like a monthly data batch), choose spot for the lowest cost. This step directly affects your monthly bill.
Launch Resources Using the AWS Free Tier Correctly
When creating a new AWS account, always launch only free tier-eligible services within their limits. For EC2, select t2.micro or t3.micro in a region that supports the free tier, and set a billing alert for £0.01 so you are notified if any charge appears. Do not assume all instances are free — check the free tier page for exact specifications.
Set Up AWS Budgets and Billing Alerts
In the AWS Management Console, navigate to Billing and Cost Management and create a budget. Set a cost budget of e.g. £10 per month. Configure an email alert at 80% and 100% of the budget. This ensures you are warned immediately if you accidentally exceed the Free Tier limits or if usage spikes unexpectedly. This step is critical to avoid an unbudgeted bill.
Monitor Usage with AWS Cost Explorer and Free Tier Usage Reports
Regularly (e.g., weekly) open AWS Cost Explorer to see a graph of your spending by service. Also check the 'Free Tier Usage' report in the Billing console, which shows exactly how much of your free allowance you have consumed (e.g., 300 hours of EC2 used out of 750 free hours). This gives you visibility to avoid exceeding limits.
Manage Resource Lifecycles and Clean Up Unused Resources
When you finish testing or a project ends, terminate all resources (like EC2 instances, RDS databases, and Elastic IP addresses) to stop further charges. Use tags (e.g., 'CostCentre: Testing') to identify resources that can be safely deleted. AWS charges for resources that are running but not actively used — this step prevents orphaned resources from silently costing money.
Optimise Costs Over Time with Right-Sizing and Reserved Instances
After a few months of running production workloads, analyse usage patterns. Use AWS Compute Optimizer to see if you are over-provisioned (e.g., using a large instance when a medium one is sufficient). Purchase reserved instances or Savings Plans for the steady-state workload. Schedule non-critical resources to stop during off-hours (e.g., stop a development server at 6pm and restart at 8am). This reduces costs without affecting performance.
Imagine you work as a junior cloud administrator for a small e-commerce company called "ShopBright". The company sells handmade candles online. You need to set up the IT infrastructure using AWS, and your boss gives you a budget of £500 per month. Every penny counts. Here is how AWS pricing models and the Free Tier factor into your real job.
Step one: You need to host the company website. The website gets steady traffic throughout the day, with occasional spikes during holiday sales. You know that on-demand EC2 instances would let you scale up during the sales, but running one instance 24/7 on on-demand pricing would cost about £30 per month for a small instance. However, you can use the AWS Free Tier for your first year — you launch a t2.micro EC2 instance (the free tier-eligible instance) and pay nothing for the first 12 months. This saves ShopBright £360 in the first year alone. But you must remember that after one year, the free tier expires, and that same instance will start costing money.
Step two: You need a database for customer orders. You predict steady, predictable usage. Reserved pricing is perfect here. You commit to a one-year term for a small Amazon RDS database instance, paying a small upfront fee (say £100) and then a lower monthly rate. Compared to on-demand pricing, reserved pricing saves ShopBright about 40% over the year. Your boss is happy because the budget stays under control.
Step three: You need to run a monthly data analysis job that processes sales data and produces a report. This job takes about two hours to run each month and is not time-sensitive. You should use a Spot Instance for this. You launch the analysis on a spot instance that costs about £0.01 per hour instead of £0.10 per hour on-demand. Over 12 months, that saves ShopBright about £2.16. It is a small amount, but it adds up across many services.
Step four: Your boss asks you to test a new AI-powered recommendation engine for the website. You have never used Amazon SageMaker before. You check the AWS Free Tier and see that SageMaker offers a two-month free trial with 250 hours of free usage. You use that free trial to test the recommendation engine thoroughly without spending a penny. After two months, you either stop using it or request budget to pay for it.
Step five: You need to store product images. Amazon S3 (Simple Storage Service) has an always free tier: 5 GB of standard storage for free. You store your images there, staying under the 5 GB limit, and pay nothing for storage. When you eventually exceed 5 GB, you pay a tiny per-GB rate.
Step six: You accidentally leave a large EC2 instance running over the weekend after testing. When you check the billing dashboard on Monday, you see an unexpected £15 charge. You immediately learn to set up AWS Budgets and billing alerts — these are free tools that send you an email when your spending exceeds a threshold you set (like £10). You also learn to tag resources (like "cost-centre: engineering") so you can track which department is spending what.
The key takeaway for your real job: you must continuously monitor usage, understand which pricing model applies to each service, and never assume something is free. The exam tests exactly these decision-making skills — given a business scenario, you choose the cheapest or most appropriate pricing model, and you remember the Free Tier limits to avoid surprise bills.
The CLF-C02 exam tests AWS Pricing Models and Free Tier heavily, because understanding cost is a core part of the Cloud Practitioner role. Expect about 5-8 questions on this topic, often woven into scenario-based questions. Here is exactly what you need to master.
The exam loves to test the following specific concepts:
The three main pricing models: on-demand, reserved, and spot. You must know the definition of each, when to use it, and the relative cost. A typical question: "A company needs to run a batch processing job that can be interrupted. Which pricing model is most cost-effective?" The answer is spot instances. If the job cannot be interrupted, the answer would be reserved or on-demand.
The difference between reserved instances and Savings Plans. The exam expects you to know that both offer discounts in exchange for a commitment, but Savings Plans are more flexible because they apply to any instance family or region within a compute category. A trap: a question says "A company wants a discount across multiple EC2 instance types but does not want to commit to a specific instance. Which should they choose?" The correct answer is Compute Savings Plan, not reserved instances.
AWS Free Tier limits. The exam tests exactly what is included. For example: "How many hours per month can you use a t2.micro EC2 instance under the Free Tier?" Answer: 750 hours per month. "What happens after the 12-month Free Tier period ends for that EC2 instance?" Answer: You start paying the on-demand rate unless you stop or terminate the instance. A common trap: the exam might say "You sign up for a new AWS account. Which EC2 instance types are free forever?" The answer is none — the free tier is only for specific instance types (t2.micro or t3.micro) and only for 12 months, not forever.
Data transfer pricing. The exam tests that data transfer INTO AWS (ingress) is usually free, but data transfer OUT of AWS (egress) costs money. A scenario question might describe a company that stores data in S3 and serves it to users on the internet. The cost includes both storage (S3) and data transfer out (egress). Many students forget the egress cost.
Consolidated billing and volume discounts. If an organisation has multiple AWS accounts (e.g., one for each department), they can link them under a single payer account. This is called consolidated billing, and it aggregates usage across all accounts, making them eligible for volume discounts. The exam tests that this does NOT require the accounts to be in the same region — volume discounts are global.
AWS Budgets and AWS Cost Explorer. The exam wants you to know that AWS Budgets lets you set spending limits and receive alerts when you exceed them. AWS Cost Explorer is a tool to visualise and analyse your past spending. These are free features of the AWS Management Console. A question might describe a scenario where a company wants to avoid surprise bills — the answer is to set up AWS Budgets with alerts.
The concept of Economies of Scale. AWS passes its cost savings (from buying massive amounts of hardware and running its own power plants) to customers in the form of lower prices over time. The exam tests that this is why cloud computing is cheaper for end users — you share the benefits of AWS's scale.
Common traps to watch for:
Trapping you on "always free" versus "12 months free". Some services (like Lambda with 1 million requests) are always free, while others (like EC2 t2.micro) are free only for 12 months. The exam might mix them up in an answer choice.
Trapping you on the exact free tier specification for EC2. The free tier covers 750 hours per month of t2.micro or t3.micro instances. If you run two instances at the same time, you use 2 hours of free tier per hour of wall clock time — you will exceed the free allowance if you run both continuously.
Trapping you on "Reserved instances are only for EC2". This is false — reserved pricing exists for RDS, ElastiCache, and other services too. The exam expects you to know that reserved pricing is available for multiple services, not just EC2.
Trapping you on the word "free". Any question that says "free tier" without specifying the service or limits should trigger caution. Always check the details in the question — is it always free, 12 months free, or a trial? Does the scenario exceed the free tier limits?
Key definitions to memorise:
On-demand: pay per hour/second, no commitment, highest cost.
Reserved: 1- or 3-year commitment, significant discount, for steady workloads.
Spot: spare capacity, up to 90% discount, can be interrupted.
Savings Plans: commit to a dollar amount per hour, flexible across services.
Free Tier: limited free usage, often for first year or forever with caps.
Consolidated billing: combine accounts for volume discounts.
AWS Budgets: set spending limits and get alerts.
Data transfer egress: money leaving AWS costs money.
AWS offers three main pricing models: on-demand (pay by the hour/second, no commitment), reserved (commit for 1-3 years for up to 72% discount), and spot (use spare capacity for up to 90% discount but can be interrupted).
The AWS Free Tier includes three types: always free (no expiry, e.g. 1 million Lambda requests per month), 12 months free (e.g. 750 hours of t2.micro EC2 per month), and short-term trials (e.g. 250 hours of SageMaker for 2 months).
Data transfer into AWS (ingress) is generally free, but data transfer out to the internet (egress) costs money — a common exam trap.
Reserved pricing applies to multiple services beyond EC2, including RDS, ElastiCache, and Redshift — do not limit it to just virtual servers.
AWS Budgets let you set spending thresholds and receive alerts to avoid surprise bills; AWS Cost Explorer helps you visualise past spending.
Consolidated billing aggregates usage across all linked AWS accounts (regardless of region) to qualify for volume discounts.
Savings Plans are more flexible than reserved instances because they apply to any instance type or region within a compute category, while reserved instances lock you to a specific configuration.
If you exceed the Free Tier limits, you automatically start paying on-demand rates — there is no warning, so you must monitor your usage.
Spot instances are ideal for fault-tolerant or batch workloads, not for mission-critical applications that cannot tolerate interruptions.
Economies of scale mean AWS passes its massive buying power to customers, making cloud computing cheaper over time than running your own data centre.
These come up on the exam all the time. Here's how to tell them apart.
On-Demand Pricing
Pay by the hour or second with no long-term commitment
Highest cost per unit of usage (e.g., per hour)
Best for unpredictable workloads or short-term projects
Reserved Pricing
Commit to a 1-year or 3-year term for a guaranteed discount
Up to 72% cheaper than on-demand rates
Best for steady, predictable workloads like a production database
Reserved Instances
Tied to a specific instance family (e.g., m5.large in one region)
Typically offers deeper discounts (higher commitment)
Less flexible — cannot change instance type without losing discount
Compute Savings Plans
Applies to any EC2 instance, Lambda, or Fargate in any region
More flexible — automatically applies to usage
Slightly lower discount than reserved instances in some cases
AWS Free Tier (12 months)
Lasts 12 months from account creation
Includes EC2 t2.micro (750 hours/month), 5GB S3, 100GB egress
Expires after 12 months — then services cost money
AWS Free Tier (Always Free)
Never expires — available indefinitely
Includes 1 million Lambda requests/month, 25GB DynamoDB, 100 cloudwatch metrics
Always free — no time limit, but still has usage caps
Spot Instances
Up to 90% cheaper than on-demand
Can be terminated by AWS with 2 minutes' notice
Only for fault-tolerant or interruptible workloads
On-Demand Instances
Full price — no discount
Guaranteed to run until you stop it
Suitable for any workload, including production
Data Transfer In (Ingress)
Generally free — no charge for uploading to AWS
Includes data sent from your office to AWS via VPN or internet
No cost consideration for most upload scenarios
Data Transfer Out (Egress)
Costs money per gigabyte (GB) transferred
Charges apply when data leaves AWS to the internet or another region
Can be a significant hidden cost for data-heavy applications
AWS Budgets
Lets you set spending limits and receive alerts
Proactive — warns you before you overspend
Can monitor free tier usage against limits
AWS Cost Explorer
Lets you visualise and analyse historical spending
Reactive — shows what you already spent
Ideal for understanding trends and identifying cost-saving opportunities
Mistake
The AWS Free Tier means everything on AWS is free for the first year.
Correct
The Free Tier only covers specific services with specific usage limits. For example, EC2 free tier is only for t2.micro or t3.micro instances up to 750 hours per month, not for all instance types or unlimited usage.
Marketing materials often highlight 'free' without emphasising the strict limits, leading beginners to launch larger instances or extra services and then get shocked by a bill.
Mistake
Reserved instances are only available for Amazon EC2 (virtual servers).
Correct
Reserved pricing is available for many AWS services including RDS (databases), ElastiCache (caching), and Redshift (data warehousing), not just EC2. Each service has its own reserved pricing offering.
The term 'Reserved Instance' is most commonly associated with EC2 in AWS documentation, so students assume it is exclusive to that service. The exam tests the broader application.
Mistake
If you use the Free Tier, you will never be charged for anything.
Correct
You will be charged if you exceed the free tier limits, use a service not covered by the free tier, or continue using services after the 12-month period ends. For example, if you use 1000 hours of EC2 in a month, you pay for the extra 250 hours.
Beginners often think 'free' is absolute, not relative to a quota. They do not realise that billing is automated, so the moment you cross the limit, charges start accumulating silently.
Mistake
Spot instances are the same as on-demand, just cheaper with no downsides.
Correct
Spot instances can be terminated by AWS with just two minutes' notice if capacity is needed elsewhere. They are only suitable for fault-tolerant or interruptible workloads, not for critical applications like a live website database.
The word 'spot' sounds like a bargain without obvious risk. Beginners do not immediately understand the trade-off between cost and reliability.
Mistake
Consolidated billing only works if all accounts are in the same AWS region.
Correct
Consolidated billing aggregates usage across all linked accounts regardless of region. Volume discounts and tiered pricing apply globally to the total usage of the organisation.
The word 'consolidated' might imply regional grouping, but AWS pricing discounts are global. This misconception leads to wrong answers in exam questions about cost optimisation.
Mistake
The AWS Free Tier includes unlimited free data transfer out to the internet.
Correct
The Free Tier includes only 100 GB of free data transfer out to the internet each month for the first 12 months, and it is shared across all services. After that or if you exceed it, you pay standard data transfer rates.
Beginners focus on compute and storage freebies but overlook data transfer, which can be a significant hidden cost. They assume data movement is free because it is free to upload data to AWS.
Reveal each answer, then mark whether you got it right. Score 60%+ to unlock the next chapter.
Set up AWS Budgets with an alert for a very low threshold (like $1) so you get an email if any charge is incurred. Also, only launch free tier-eligible resources and check the Free Tier Usage report in the console regularly to see how much of your allowance you have used.
The Free Tier includes 750 hours per month of t2.micro or t3.micro instances for 12 months from account creation. That is enough to run one instance continuously for the whole month. If you run two instances, you use 1,488 hours in a month and will be charged for the extra 738 hours at the on-demand rate.
On-demand means you pay by the hour or second with no commitment — it is flexible but expensive. Reserved means you commit to using a service for 1 or 3 years and get a significant discount (up to 72%) in return. Reserved is ideal for predictable, steady workloads.
No, spot instances are not suitable for a live website because AWS can reclaim the capacity with only two minutes' notice, which would cause your site to go offline. Spot instances are best for fault-tolerant or batch workloads that can handle interruptions, like data processing or rendering.
After 12 months, you lose the free allowance for that service. For example, your t2.micro EC2 instance will start being billed at the on-demand rate unless you stop or terminate it. Always set a calendar reminder to review your resources before the 12-month mark.
No. Data transfer into AWS (uploading) is generally free, but data transfer out to the internet (downloading) costs money per gigabyte. The Free Tier includes 100 GB of free egress per month for the first 12 months. Always factor egress costs into your budget for applications that serve data to users.
Consolidated billing lets you link multiple AWS accounts under one paying account. It aggregates usage across all accounts, which qualifies the entire organisation for volume discounts and tiered pricing. It does not matter if the accounts are in different regions — the discounts apply globally.
You've just covered AWS Pricing Models and Free Tier — now see how well it sticks with free CLF-C02 practice questions. Full explanations included, no account needed.
Done with this chapter?