Cost Explorer, Budgets, and Cost Allocation Tags

Without visibility into your AWS spending, cost optimisation is impossible. You cannot improve what you cannot measure. AWS provides three primary tools for cost visibility: Cost Explorer for analysing historical and forecasted spend, AWS Budgets for proactive alerts when spending exceeds thresholds, and Cost Allocation Tags for breaking down costs by team, project, or environment. Together, these tools give you the financial management foundation the Well-Architected Cost Optimisation pillar requires.

AWS Cost Explorer provides a visual interface and API to explore your AWS cost and usage data. You can view spend by service, region, account, resource type, tag, and time period. Cost Explorer also provides 12-month spend forecasts based on historical trends, RI and Savings Plan coverage reports, and cost anomaly detection that uses ML to alert you when spending patterns change unexpectedly. Enable Cost Explorer in the Billing console — it takes 24 hours to populate data.

AWS Cost Anomaly Detection uses machine learning to continuously monitor your costs and alert you to unexpected increases. It learns your normal spending patterns and detects statistically significant deviations. You configure anomaly monitors (what to monitor: service, linked account, cost category, or tag) and alert subscriptions (SNS email or Slack webhook when anomalies exceed a threshold). Anomaly Detection is invaluable for catching runaway Lambda costs, accidental large EC2 instance launches, or unexpected data transfer charges before they escalate.

AWS Budgets lets you set custom spending thresholds and receive alerts when actual or forecasted spend crosses them. Four budget types: Cost Budget (dollar amount), Usage Budget (hours, GB, requests), RI Utilisation Budget (alert if utilisation drops below threshold), and Savings Plans Utilisation Budget. Configure multiple alert thresholds per budget (e.g., 50%, 80%, 100% of budget, plus 110% of actual). Notifications go to email or SNS, and Budget Actions can automatically apply SCPs or IAM policies to stop further spending.

AWS Budget Actions can automatically respond when a budget threshold is breached — no human required. You can configure actions to: Apply an IAM policy that denies the ability to launch new EC2 instances. Apply a Service Control Policy (SCP) to an AWS Organizations target. Stop EC2 or RDS instances that are over budget. For example, a development account budget can automatically stop all EC2 instances if the monthly budget exceeds $5,000, preventing runaway costs from forgotten test environments.

Cost Allocation Tags enable you to attribute AWS costs to specific business dimensions. Two types: AWS-generated tags (e.g., aws:createdBy) and user-defined tags (your custom key-value pairs). You must activate tags in the Billing console before they appear in cost reports. Once activated, tags flow into Cost Explorer, CUR, and Budgets for filtering. Define a tag taxonomy: Project, Environment, Team, CostCenter — enforce it with Tag Policies in AWS Organizations.

Tags are only useful if consistently applied. Tag Policies in AWS Organizations enforce tag key and value formats across all accounts in an organisation. A tag policy can specify that the Environment tag must exist and must be one of Production, Staging, or Development. Non-compliant resources appear in the tag policy compliance report. Combine tag policies with AWS Config rules (e.g., required-tags) to detect and alert on untagged resources before they accumulate costs you cannot attribute.

The AWS Cost and Usage Report (CUR) is the most granular billing data available — hourly line items for every resource in your account, including resource IDs, tags, Savings Plan coverage, and blended/unblended costs. CUR data is delivered to an S3 bucket in CSV or Parquet format. Query it with Amazon Athena for custom analysis, or feed it into Amazon QuickSight for dashboards. CUR is essential for organisations that need fine-grained showback or chargeback to business units.

AWS Cost Categories let you map cost and usage into meaningful business categories using rules. For example, you can create a category called Production Environment that includes all resources tagged Environment=Production plus specific account IDs. Categories appear in Cost Explorer as a filter dimension, making it easy to see total production vs development costs even if your tagging is inconsistent. Cost Categories support rule-based mapping, so they can compensate for incomplete or inconsistent tagging historically.

FinOps (Financial Operations for the cloud) is the practice of managing cloud costs as a team responsibility, not just a finance function. Key practices: Visibility — every team can see their own spending in Cost Explorer. Accountability — teams own their cost allocations and budgets. Optimisation — regular right-sizing and RI/SP review. Forecasting — predict future spend using Cost Explorer forecasts. On AWS, implement FinOps by creating per-team AWS accounts under Organizations, per-team budgets, tag policies, and monthly cost reviews with engineering leads.

Cost Explorer forecasting uses your historical usage patterns to predict future costs up to 12 months out. This helps with budget planning and capacity planning. For new workloads with no history, use the AWS Pricing Calculator to estimate costs before deployment. Always add a 20-30% buffer to your forecast to account for unexpected growth or new services. For organisations with complex cost structures, consider AWS Budgets** reports that automatically generate monthly cost summaries delivered to email or S3.

Test your understanding of AWS Solutions Architect (SAA-C03) concepts from this lesson.

In this lesson you learned: Cost Explorer analyses historical and forecasted spend with cost anomaly detection, AWS Budgets provides proactive alerts and automated Budget Actions when thresholds are exceeded, and Cost Allocation Tags enable showback and chargeback by attributing costs to teams, projects, and environments. Tag Policies in AWS Organizations enforce consistent tagging. Next up we explore S3 and data transfer cost optimisation techniques.