Billing, Pricing, and Support practice questions

Reserved Instances (Standard)

Why wrong: Standard Reserved Instances are tied to a specific instance family (e.g., M5) and region. Switching families would require purchasing new RIs and forfeiting the discount on the original RIs, so this does not provide the desired flexibility.

EC2 Instance Savings Plans

Why wrong: EC2 Instance Savings Plans are similar to Standard Reserved Instances in that they are limited to a specific instance family within a region (e.g., M5 in US-East-1). They do not allow changing families, so they fail to meet the flexibility requirement.

Spot Instances

Why wrong: Spot Instances provide substantial discounts but are interruptible—AWS can reclaim capacity with only a two-minute warning, making them unsuitable for workloads that require consistent, guaranteed availability. They are best suited for fault-tolerant, stateless, or time-insensitive batch jobs, not for the company's need to run multiple workloads on a predictable, long-term basis. Additionally, Spot pricing and capacity vary over time, so they do not offer the reliable cost savings of a committed-plan discount.

AWS Cost Explorer with a cost allocation tag filter

Why wrong: AWS Cost Explorer is a read-only analytics service that visualizes historical and forecasted AWS costs and usage, with filtering capabilities such as cost allocation tags to group spend by project or environment. It cannot execute automated actions like stopping EC2 instances when a cost threshold is exceeded, as it lacks any enforcement or automation engine. While Cost Explorer helps identify where costs are accumulating, it does not provide the intervention mechanism required for this scenario.

AWS Trusted Advisor with the cost optimization check

Why wrong: AWS Trusted Advisor provides recommendations to optimize costs, such as identifying idle instances, but it does not automatically take actions like stopping instances when a cost limit is exceeded. It is a guidance tool, not an enforcement tool.

Amazon CloudWatch with a billing metric alarm

Why wrong: Amazon CloudWatch can monitor billing metrics (e.g., estimated charges) and trigger an alarm, but the alarm cannot directly stop EC2 instances. It can send notifications via SNS, but automated stopping would require additional custom integration (e.g., Lambda), which is not the native feature described.

AWS Developer Support

Why wrong: The Developer Support plan provides best-effort response times during business hours and does not include a designated Technical Account Manager (TAM) or 15-minute response time. It is suitable for development and test environments but not for critical production workloads requiring rapid support.

AWS Business Support

Why wrong: The Business Support plan offers a 1-hour response time for production system impaired cases, which is slower than the required 15 minutes. It does not include a designated Technical Account Manager (TAM); guidance is provided by AWS support engineers. It is not the highest level of support.

AWS Enterprise On-Ramp

Why wrong: Enterprise On-Ramp provides a designated Technical Account Manager (TAM) but offers a 1-hour response time for production system impaired cases, not 15 minutes. It is designed for customers who are building a long-term relationship with AWS but may not yet require the full Enterprise plan.

AWS Cost Explorer

Why wrong: AWS Cost Explorer provides interactive charts and graphs to visualize cost and usage data, and it can generate reports that are emailed on a schedule. However, it does not deliver the most granular raw line-item CSV to an S3 bucket automatically. Cost Explorer's downloadable reports are aggregated and less detailed than the Cost and Usage Report.

AWS Budgets

Why wrong: AWS Budgets is used to set custom cost and usage budgets and receive alerts when thresholds are exceeded. It does not generate or export raw cost and usage data files for BI tool integration. Its purpose is monitoring and notification, not data export.

AWS Trusted Advisor

Why wrong: AWS Trusted Advisor inspects your AWS environment and provides best practice recommendations across cost optimization, performance, security, and fault tolerance. It does not generate or export raw cost and usage data. It offers cost-saving suggestions but not the raw data feed required for BI tool integration.

Basic Support

Why wrong: The Basic Support plan includes only documentation, whitepapers, and the AWS Support forum. It does not provide a dedicated TAM, fast response times for business-critical issues, or Infrastructure Event Management. It is not suitable for a company requiring proactive architecture guidance and rapid response for critical workloads.

Developer Support

Why wrong: The Developer Support plan offers technical support via email during business hours, with a response time of up to 12 hours for critical issues. It does not include a dedicated TAM, 15-minute response for business-critical issues, or Infrastructure Event Management. It is intended for early development and testing, not for production-critical systems requiring proactive guidance.

Business Support

Why wrong: The Business Support plan provides 24/7 access to Cloud Support Engineers, a 1-hour response time for business-critical issues, and support for third-party software. However, it does not include a dedicated Technical Account Manager (TAM) or Infrastructure Event Management. While better than Developer, it still lacks the personalized, proactive architectural guidance and the 15-minute response time that the company requires.

AWS Cost Explorer

Why wrong: AWS Cost Explorer provides rich dashboards and graphs to visualize cost and usage data, but it does not send proactive alerts based on budget thresholds or cost anomalies. It is a reactive analysis tool, not an alerting service.

AWS Trusted Advisor

Why wrong: AWS Trusted Advisor performs best-practice checks and delivers recommendations for cost optimization, performance, security, and fault tolerance. While it can flag underutilized resources or suggest savings, it does not monitor real-time cost usage against user-defined budget thresholds, nor does it employ machine learning to detect anomalous daily cost spikes. Its cost-related insights are advisory and non-alerting for these specific use cases.

AWS CloudTrail

Why wrong: AWS CloudTrail records API activity and management events across accounts, enabling governance, compliance, and operational auditing. It does not ingest or process cost and usage data, nor does it evaluate spend against budgets or detect anomalies in daily billing patterns. Therefore, it is unrelated to the requirement to alert on cost thresholds or unexpected spending spikes.

EC2 Instance Savings Plans (3-year, no upfront)

Why wrong: EC2 Instance Savings Plans are tied to a specific instance family in a region (e.g., m5 in us-east-1). They do not allow moving to a different instance family or to services like Fargate or Lambda, so they do not meet the flexibility requirement.

Standard Reserved Instances (1-year, all upfront)

Why wrong: Standard Reserved Instances are specific to a particular instance type, size, and Availability Zone. They cannot be used across different instance families or non-EC2 services, and the 1-year term does not match the desired 3-year commitment for maximum savings.

Convertible Reserved Instances (3-year, all upfront)

Why wrong: Convertible Reserved Instances offer flexibility to change instance attributes, but they are limited to EC2 instances and require an exchange process that may not always be available. They do not cover Fargate or Lambda usage, so they are less compatible with the company's multi-service environment.

Set the 'shared' attribute to 'true' when purchasing the Reserved Instance.

Why wrong: A Reserved Instance purchase request accepts parameters such as instance type, tenancy, platform, and scope (Regional or Availability Zone), but there is no 'shared' attribute or any equivalent flag in the purchase workflow. The visibility of a Reserved Instance across accounts is controlled exclusively by the organization-level 'Reserved Instance sharing' setting in the AWS Billing and Cost Management console. This setting is applied after a purchase and can be toggled at any time; it is not a per-purchase field, so attempting to set such an attribute during purchase is not technically possible and would have no effect.

Enable Reserved Instance sharing in the AWS Management Console under the Billing and Cost Management service.

Why wrong: This is the correct action. The Reserved Instance sharing setting is found in the Billing and Cost Management console. Enabling it allows the member account's RI discount to be applied to eligible usage across all accounts in the organization, provided the aggregated usage meets the RI requirements.

Enable the Reserved Instance sharing setting in the AWS Organizations console.

Why wrong: The Reserved Instance sharing setting is not available in the AWS Organizations console. It is configured in the Billing and Cost Management console. The Organizations console is used for managing policies, accounts, and organizational units, not RI sharing.

Compute Savings Plan

Why wrong: Compute Savings Plans apply to EC2 instance usage regardless of family, size, region, or tenancy, and also cover AWS Fargate and AWS Lambda compute. This broad architectural flexibility is the reason they carry the lowest possible discount among Savings Plans, even lower than EC2 Instance Savings Plans. For a steady-state workload running on a specific EC2 instance type, you would leave potential savings on the table by choosing a Compute Savings Plan instead of a Standard Reserved Instance, which locks down the exact configuration.

Convertible Reserved Instances (3-year, All Upfront)

Why wrong: Convertible Reserved Instances allow you to change instance families, operating systems, and tenancies over the term, which is a valuable flexibility feature. However, that flexibility comes at a cost: AWS applies a lower discount rate than Standard Reserved Instances because the commitment is less rigid. Even with a 3-year, All Upfront payment, a Convertible RI will not match the maximum savings that a Standard RI achieves for a fixed, predictable production workload.

Standard Reserved Instances

Why wrong: Standard Reserved Instances provide the largest discount and offer a capacity reservation, but they are locked to a specific instance family, operating system, tenancy, and Region. While you can modify the Availability Zone, instance size within the same family, or network type, the management console does not allow changing the instance family or OS, nor can you exchange them for a different Region. These constraints eliminate Standard Reserved Instances as an option for the described cross-Region flexibility.

Scheduled Reserved Instances

Why wrong: Scheduled Reserved Instances are designed for workloads that run on a recurring schedule, such as daily batch jobs or predictable time windows, rather than for broad on-demand flexibility. Each reservation is pinned to a specific instance type, Region, and schedule, and you cannot modify the instance family, OS, tenancy, or move the reservation to another Region. Because the question asks for the ability to change instance family, OS, and tenancy, Scheduled Reserved Instances are not a valid answer.

Compute Savings Plans

Why wrong: Compute Savings Plans offer flexibility in instance family, size, OS, and tenancy within a single AWS Region, but the commitment is fixed to that Region and cannot be applied in another Region. They do not provide a capacity reservation, and there is no exchange mechanism to move the plan across Regions. Therefore, although they meet the flexibility requirements for compute attributes, they fail the cross-Region requirement stated in the question.

Developer Support

Why wrong: Developer Support is intended for development and testing environments, not production workloads. It provides email access only during business hours, with a target response time of 12 hours for impaired systems and no 24/7 phone or chat, so it could not meet a 1-hour production outage SLA. This tier is above Basic Support but still insufficient for business-critical workloads.

Basic Support

Why wrong: Basic Support is included at no additional cost with every AWS account, but it offers only documentation, whitepapers, and community forums, with no direct access to Cloud Support Engineers. Account and billing questions are handled separately, and users cannot open technical support cases, so there is no response-time commitment for production incidents. Therefore, it is the lowest tier and clearly not the minimum plan for a 1-hour production down response.

Enterprise On-Ramp

Why wrong: Enterprise On-Ramp adds a pool of Technical Account Managers (TAMs) and provides a 30-minute response for business-critical system down incidents, which is faster than Business Support's 1-hour production down response. However, it is a higher-priced, more advanced tier designed for organizations that need additional architectural and operational guidance, not the minimum plan that offers a 1-hour SLA. Because the question asks for the minimum plan, Enterprise On-Ramp cannot be the correct choice.

Business Support

Why wrong: Business Support is a reactive support tier that provides access to Cloud Support Engineers 24/7 for break-fix issues, with response times up to 1 hour for production-impacting incidents. However, it does not include any proactive monitoring or AWS-initiated incident response capabilities. Incident Detection and Response is a specialized add-on that is only offered with Enterprise Support, so choosing Business Support would leave you without the automated anomaly detection and AWS-managed response that the feature provides. Therefore, Business Support is incorrect for this scenario.

Developer Support

Why wrong: Developer Support is primarily designed for development and testing workloads, offering business-hours access to support resources and guidance for general architectural questions. It lacks 24/7 severity-based response coverage for production incidents and has no proactive incident management or monitoring features. Because Incident Detection and Response relies on continuous telemetry analysis and AWS-engineer-initiated actions, it cannot be delivered through this tier, which is strictly oriented toward reactive, non-critical development assistance. Thus, Developer Support does not meet the requirements of the add-on.

Business Support

Why wrong: Business Support provides 24/7 phone support and less than 1 hour response for critical issues, but it lacks Infrastructure Event Management and a designated Technical Account Manager (TAM).

Developer Support

Why wrong: Developer Support lacks 24/7 phone support, Infrastructure Event Management, and a designated Technical Account Manager (TAM).

Enterprise On-Ramp

Why wrong: Enterprise On-Ramp provides 24/7 phone support, less than 1 hour critical response, and Infrastructure Event Management, but it does not include a designated Technical Account Manager (TAM), so it does not meet all requirements.

AWS Cost Explorer

Why wrong: AWS Cost Explorer provides visualizations and analysis of your AWS costs and usage over time, but it does not natively send proactive email alerts when predefined thresholds are breached. It can save custom reports for periodic review, but it lacks the automated notification capability required for proactive monitoring.

AWS Trusted Advisor

Why wrong: AWS Trusted Advisor is a service that inspects your AWS environment and makes recommendations for cost optimization, performance, security, and fault tolerance. While it can identify underutilized resources, it does not allow you to set custom spending thresholds or send email alerts when costs reach a specific dollar amount.

AWS Consolidated Billing

Why wrong: AWS Consolidated Billing is a feature of AWS Organizations that aggregates costs from multiple accounts into a single bill, enabling volume discounts. It does not provide threshold-based alerting. To set budget alerts across consolidated accounts, you would use AWS Budgets, not Consolidated Billing itself.

AWS Cost Explorer

Why wrong: AWS Cost Explorer is incorrect for this requirement because it is a cost and usage analysis service that visualizes historical and forecasted spending through pre-built or custom reports, such as daily/monthly cost breakdowns by service or linked account. It operates on top of existing billing data but does not alter how accounts are invoiced, nor does it aggregate usage across accounts into a single bill. At best, it helps you filter consolidated data after billing has already been combined.

AWS Budgets

Why wrong: AWS Budgets is an incorrect choice because it is a proactive monitoring and alerting tool, not a billing consolidation mechanism. It lets you define custom spending thresholds and receive notifications via email or Amazon SNS when your actual or forecasted costs exceed those limits, but it does not combine the usage of multiple accounts into one invoice. While it can track overall spend, the underlying billing relationship between accounts remains unchanged.

AWS Trusted Advisor

Why wrong: AWS Trusted Advisor is not the right answer because it is an advisory and optimization service that evaluates your environment against six pillars, including cost optimization, security, and fault tolerance, and provides recommendations for improvement. It cannot aggregate usage or issue a consolidated invoice, as its role is to suggest actions such as rightsizing underutilized resources or deleting idle resources. It may support cost reduction, but that is a recommendation, not a billing mechanism.

AWS Budgets

Why wrong: AWS Budgets is fundamentally a proactive alerting service that lets you define cost or usage thresholds and receive notifications when your actual or forecasted spend exceeds those limits. It is designed to help you stay within a predetermined budget, not to enable historical exploration of past costs or render charts breaking down spend by service, linked account, or region over a multi-month period. Budgets can show the current status against a target, but it does not provide the ability to look back six months and analyze trends or anomalies. Consequently, it is not an appropriate tool for the retrospective spending review described in the question.

AWS Trusted Advisor

Why wrong: AWS Trusted Advisor performs real-time inspection of your environment and generates recommendations based on current service usage and configuration, such as identifying idle RDS instances or underutilized EIPs. These checks are point-in-time assessments focused on cost optimization, security, performance, and fault tolerance, not retrospective analyses. Trusted Advisor does not let you query or filter historical spend across a custom date range, nor does it provide interactive dashboards for visualizing cost trends over the past six months. Therefore, while it can uncover savings opportunities at the moment it runs, it cannot meet the requirement to review and analyze six months of spending history.

AWS Cost and Usage Reports

Why wrong: AWS Cost and Usage Reports deliver the most detailed billing data as CSV files, but they require additional tools (e.g., Amazon Athena, Amazon QuickSight) for querying and visualization. They are not designed for immediate interactive exploration.

AWS Developer Support

Why wrong: AWS Developer Support is intended for development and test workloads, not production. Its fastest (Severity 1) response target is 12 hours, which is far slower than the required under-1-hour urgent response. In addition, Trusted Advisor access is limited to the 7 core checks, not full best-practice checks. Therefore Developer Support is insufficient for a critical production application and cannot be the lowest-cost qualifying option.

AWS Enterprise On-Ramp Support

Why wrong: The Enterprise On-Ramp plan provides a 30-minute response for urgent issues and full Trusted Advisor access. However, it is more expensive than the Business plan and includes additional features not required, making it not the lowest-cost option that meets the stated requirements.

AWS Enterprise Support

Why wrong: The Enterprise Support plan offers a 15-minute response for urgent issues and full Trusted Advisor access. It is the most expensive plan and includes features not needed for the requirement, so it is not the lowest-cost option.

AWS Budgets

Why wrong: AWS Budgets allows you to set custom cost and usage budgets and receive alerts when thresholds are exceeded, but it does not consolidate billing across accounts or enforce permissions such as restricting regions.

AWS Identity and Access Management (IAM) cross-account roles

Why wrong: IAM cross-account roles enable users in one account to assume roles in another account for temporary access, but they do not provide a way to consolidate billing or enforce blanket restrictions like a region lock across all accounts. Setting up such restrictions would require manual configuration in each account and does not offer central enforcement.

AWS Cost and Usage Reports

Why wrong: AWS Cost and Usage Reports provide detailed cost and usage data that can be analyzed with tools like Amazon Athena, but they do not consolidate billing or provide any policy enforcement capabilities. They are a reporting tool, not a governance or billing consolidation service.

AWS Cost Explorer with a custom cost threshold report

Why wrong: Incorrect. AWS Cost Explorer provides interactive dashboards and reports for analyzing cost and usage data, but it does not automatically take corrective actions like stopping instances. It is a visualization and analysis tool, not an automation tool.

AWS Trusted Advisor with a cost optimization check

Why wrong: AWS Trusted Advisor with a cost optimization check is incorrect because Trusted Advisor is an advisory service that inspects your AWS environment and provides recommendations, such as identifying idle EC2 instances or underutilized resources. It does not monitor real-time spending thresholds, and it cannot be configured to automatically stop EC2 instances when a cost limit is exceeded. While Trusted Advisor offers a cost optimization check, it is purely informational and requires manual intervention to act on its findings, unlike AWS Budgets with budget actions, which automates the stop action natively.

AWS Organizations with a service control policy (SCP)

Why wrong: Incorrect. Service control policies (SCPs) in AWS Organizations allow you to centrally control permissions for member accounts, such as restricting the launch of certain instance types or services. They do not monitor costs or trigger actions based on spending thresholds.

EC2 Instance Savings Plans

Why wrong: EC2 Instance Savings Plans are incorrect for this scenario because they lock the discount to a specific instance family in a specific AWS Region, such as only m5.large in us-east-1. Although they provide deeper discounts than Compute Savings Plans, they do not cover AWS Fargate usage at all, which is a critical gap since the company runs both EC2 and Fargate. This plan is too restrictive for a mixed environment where containers and virtual machines may need to change over time, and it would force separate commitments or break cost coverage if resources change families.

Reserved Instances (Convertible)

Why wrong: Convertible Reserved Instances are incorrect because they apply only to EC2 instances and have no applicability to AWS Fargate, even though they allow exchanging one instance attribute for another over the term. While Convertible RIs provide flexibility to change instance families, operating systems, or tenancies via exchange, that flexibility is still confined to the EC2 service itself. Since the company runs Fargate containers in addition to EC2, selecting Convertible RIs would leave all container compute at On-Demand prices, failing to achieve the broad cost reduction the CFO wants across the entire mixed compute environment.

On-Demand Instances

Why wrong: On-Demand Instances are incorrect because they involve no upfront commitment and therefore offer no volume-based discount or savings benefit from committing to a consistent level of usage. The CFO explicitly wants to reduce compute costs through a commitment, but On-Demand pricing charges the full list rate for every second of EC2 or Fargate usage. Without a Savings Plan or Reserved Instance, the company would miss out on the significant cost reductions available for predictable compute workloads, so this option fails the stated requirement.