Our methodology: how CloudMetrics compares cloud prices

Data sources

CloudMetrics retrieves pricing data directly from each cloud provider's official pricing APIs and public pricing endpoints. We do not use third-party aggregators or scraped data. Our sources are:

• AWS: The AWS Price List API and Bulk API, which provides machine-readable JSON pricing for all EC2 instance types across all regions.
• Azure: The Azure Retail Prices API, which returns pricing for all Azure services including Virtual Machines.
• GCP: The Cloud Billing Catalog API and published machine type pricing for Compute Engine.

Update frequency

Our pricing data is refreshed automatically on a regular schedule. Cloud providers typically update their public pricing monthly for standard instances, though spot/preemptible prices fluctuate more frequently. Our system:

• Fetches updated on-demand and reserved pricing data regularly.
• Validates new data against previous values to detect anomalies.
• Displays the last update timestamp on the comparison tool for transparency.
• Caches validated pricing to ensure fast response times.

Important note: while we strive for accuracy, prices shown are estimates. Your actual costs may vary based on specific agreements, enterprise discounts, or marketplace commitments with each provider.

Instance equivalence matching

Comparing instances across providers isn't straightforward. A "4 vCPU, 16 GB RAM" instance on AWS isn't identical to a similarly specced one on Azure or GCP — processors differ, network bandwidth varies, and storage options are different.

Our equivalence matching algorithm works as follows:

1. Resource specification: you specify your requirements (vCPUs, memory, storage). We search each provider's catalog for instances that meet or exceed these requirements.
2. Family categorization: we group instances into categories (general purpose, compute optimized, memory optimized, burstable) to ensure like-for-like comparisons.
3. Best-fit selection: for each provider, we select the instance that most closely matches your requirements while minimizing over-provisioning.
4. Price normalization: all prices are shown in USD per month (730 hours), making direct comparison straightforward.

Storage cost calculation

When you specify storage requirements, we calculate costs using each provider's block storage pricing:

• AWS: Amazon EBS (gp3 for SSD, st1 for HDD) pricing per GB-month.
• Azure: Managed Disk pricing (Premium SSD, Standard HDD) per GB-month.
• GCP: Persistent Disk pricing (SSD, Standard) per GB-month.

Storage prices are added to compute prices to show total monthly cost — a more realistic view of your actual expenses.

Regional pricing

Cloud pricing varies significantly by region. Our tool allows you to select specific regions for each provider independently, reflecting the reality that organizations often use different regions across providers.

We support all publicly available regions for each provider, including newer regions that may have limited instance type availability. If a specific instance type isn't available in your selected region, we indicate this clearly rather than showing an incorrect price.

What we don't include

For clarity and honesty, here's what our comparisons do not account for:

• Enterprise discount agreements: custom pricing negotiated directly with providers.
• Data transfer costs: egress charges vary significantly by provider and usage pattern.
• Support plan costs: enterprise support adds 3–10% to your total bill.
• Management overhead: multi-cloud architectures require additional tooling.
• Performance differences: two instances with identical specs may perform differently due to processor architecture.

Feedback and corrections

If you notice a pricing discrepancy or have suggestions for improving our methodology, please contact us. We take data accuracy seriously and investigate any reported issues promptly.