{
  "id": 2124804,
  "title": "AWS Trusted Advisor vs Compute Optimizer vs Cost Explorer: How Much Recoverable Waste Each One Actually Finds",
  "url": "https://urgent.news/2026/08/20/aws-trusted-advisor-vs-compute-optimizer-vs-cost-explorer-how-much",
  "topic": "tech",
  "section": "Tech",
  "published": "2026-08-20T11:27:36.000Z",
  "source": {
    "name": "Dev.to",
    "slug": "dev-to",
    "url": "https://dev.to/zop_8abedcc7e12/aws-trusted-advisor-vs-compute-optimizer-vs-cost-explorer-how-much-recoverable-waste-each-one-3g4n"
  },
  "original_language": "en",
  "account": "AWS provides three tools for cost optimization: Trusted Advisor, Compute Optimizer, and Cost Explorer. While they share some overlapping names and functions, their specific purposes differ, and their coverage varies.\n\nTrusted Advisor primarily functions as a checklist engine, running point-in-time checks for low-utilization EC2 instances, idle load balancers, unassociated Elastic IPs, underutilized EBS volumes, idle RDS instances, and other similar issues. However, the full cost checks require a Business or Enterprise support plan. It identifies known-bad patterns but does not provide dollar rollups that can be handed to finance.\n\nCompute Optimizer is a rightsizing recommender that applies machine learning to utilization history - defaulting to 14 days, up to about three months with enhanced infrastructure metrics. It suggests target configurations for EC2, ASGs, EBS, Lambda, and some container workloads. Compute Optimizer is free, specific, and focuses solely on compute resources, never addressing other categories such as NAT gateways or unattached volumes.\n\nCost Explorer is a spend analytics tool with two recommendation surfaces: reservation and Savings Plans purchase recommendations based on your usage curve, and a basic rightsizing view that overlaps with Compute Optimizer's territory but with less depth. Cost Explorer answers questions about where your money is going and what you should commit to, but it does not address waste-related issues.\n\nWhen comparing these tools against a comprehensive waste taxonomy, gaps become apparent. The eight categories that cover most recoverable spend include idle resources, rightsizing, scheduling, orphans, commitment coverage, storage class optimization, NAT gateway processing charges, and Kubernetes workload waste. Native tooling sees only a fraction of this taxonomy, often limited to the easier categories and those tied to compute resources. The categories with the biggest dollar values, such as scheduling, orphans at scale, and Kubernetes, are structural blind spots because each tool is scoped to its specific question.\n\nTo measure the coverage of these tools for your own account, it's recommended to write down the eight categories and run all three tools, exporting every finding. Then, bucket each finding into a category and calculate the monthly dollar value. Additionally, perform a manual sweep of two additional categories - scheduling opportunities and orphan queries - to compare the total dollar values surfaced by the tools versus the total waste. This comparison will give you a personal coverage number, which varies based on your architecture. Most teams find that the trio of tools excels at commitment math and basic rightsizing, while the majority of recoverable dollars lie in the unchecked categories. This information provides a defensible answer to the question of whether the provider believes your account is already optimized.",
  "summary": "Ask AWS how to cut your bill and you get pointed at three tools with overlapping names and different jobs: Trusted Advisor, Compute Optimizer, and Cost Explorer. Most comparisons list features. The question that actually matters is coverage: of the money you could realistically recover this quarter, how much can each tool even see? That framing changes the conclusion. The tools are good at what…",
  "key_points": [],
  "editors_take": null,
  "illustration": null,
  "coverage": {
    "outlets": 1,
    "also_reported_by": []
  },
  "ai_generated": true,
  "disclaimer": "Summaries, key points and the editor’s take are written by software from other outlets’ reporting and may contain errors — always check the linked original."
}