Before logs outspend the app
CloudWatch can outspend the app — estimate the stack before it does
Roll up logs, metrics, Insights, Container Insights, and Application Signals / X-Ray traces into one monthly number — and see what 7 / 30 / 90 day retention does to the bill. AMP ingest, AMG seats, and managed collector OTLP lines are out of scope — use the CloudWatch calculator or the July 2026 collectors post.
- No signup
- Instant results
- ~3 min
Rates reviewed
Logs & metrics
us-east-1list rates, pricing as of 2026-10-17. Source: AWS public pricing.
Ingest is usually the largest CloudWatch line item.
Billed like standard custom metrics ($0.30 first tier); only high-resolution alarms cost more.
Insights, containers & traces
Container Insights with OpenTelemetry for EKS: $0.08/GB, planned at ~9 GB per node per month (AWS Example 24).
Your Observability Stack Cost
Logs + metrics + traces
Estimated Monthly Stack Cost
live resultCost Breakdown
Line items and retention what-ifs unlock after email
See the Full Observability Breakdown
Per-meter line items plus 7/30/90 day retention scenarios — emailed to you.
Breakdown unlocked!
Stack line items and retention what-ifs below.
Line Items
Retention What-If
Quick wins
- Drop verbose DEBUG in production; sample high-cardinality logs.
- Set retention by log group — not one global 90-day default.
- Prefer EMF / custom metrics over parsing every log line in Insights.
- Sample Application Signals / X-Ray traces; 100% tracing gets expensive fast.
- Scraping
/metricsinto CloudWatch? Since Jul 31, 2026, managed collectors bill OTLP GB + collector-hours — model in the CloudWatch calculator.
Want this applied to your real account?
Observability bills surprise teams after a traffic spike or a chatty sidecar. We tune retention, sampling, and metric cardinality against CUR — often reclaiming more than the monitoring tool itself costs.
Talk to FinOps Experts →Who This Tool Is For
DevOps and FinOps leads whose CloudWatch line item keeps climbing — or who are sizing observability before a launch.
Why We Built This Tool
Logs ingest at $0.50/GB dominates most stacks. Metrics, Insights scans, Container Insights, and Application Signals / X-Ray traces stack on top. This roll-up uses us-east-1 list rates as of 2026-10-17 so you can budget before the invoice. AMP ingest, AMG seats, and managed Prometheus collector OTLP/collector-hour meters are not modeled here — see the CloudWatch calculator and the <a href="/blog/cloudwatch-managed-prometheus-collectors-july-2026/" class="font-semibold text-primary hover:underline">July 2026 collectors post</a>.
What Problem It Solves
- Logs share warning. Flagged when ingest+storage exceed ~60% of the stack.
- Retention what-if. 7 / 30 / 90 day totals side by side.
- More than CloudWatch metrics. Container Insights and Application Signals / X-Ray traces included.
- Free tier toggle. Model first-account vs established accounts.
Drill into the CloudWatch calculator, the AMG workspace guide for seats, or the security compliance stack for GuardDuty/Config roll-ups.
Frequently Asked Questions
How is log storage estimated?
Steady-state storage ≈ daily ingest × retention days. Actual storage depends on compression and how evenly ingest arrives.
What is the Container Insights rate?
Container Insights with OpenTelemetry for EKS bills $0.08 per GB ingested, with no separate metric, API, or storage charge. AWS pricing Example 24 shows a 50-node, 200-pod cluster at 450 GB/month ($36), so this tool plans ~9 GB per node per month. Pod count and label volume move the real figure. Enhanced observability (non-OTel) instead bills $0.21 per million observations — about $51.75/month for AWS's 10-node example.
How are Application Signals / X-Ray traces priced here?
Uses the classic X-Ray planning rate of $5.00 per million traces after the first 100,000 free traces per month (when free tier is enabled). Unified Application Signals / Transaction Search span pricing can differ — verify on the CloudWatch pricing page for your span volume.
Does this include AMP or Amazon Managed Grafana?
No. AMP ingest and AMG seats are out of scope. See the Amazon Managed Grafana 2026 guide and the observability-cost-model.csv artifact for those line items.
Does this include CloudWatch managed Prometheus collectors (July 2026)?
No — OTLP metric GB ($0.50/GB), collector-hours ($0.04/hr), and samples-collected fees are modeled in the CloudWatch pricing calculator and the managed collectors blog post with the OTLP vs classic cost CSV.
Are third-party APM tools included?
No — this models native AWS meters only. Datadog, New Relic, and similar are separate contracts.
Do high-resolution metrics cost more?
No. AWS bills high-resolution custom metrics in the same tiers as standard custom metrics ($0.30 for the first 10,000). High-resolution alarms do cost more — $0.30 per alarm metric versus $0.10 standard.
Is the Application Signals free tier included?
No. Application Signals has a 3-month free tier per account (100 GB ingested with transaction spans, 1M spans indexed, or 100M signals without spans). This tool prices traces with the classic X-Ray rate only.
How accurate is this?
Matches published CloudWatch list arithmetic for your inputs as of 2026-10-17. Cross-region, vended logs discounts, Application Signals span meters, and Contributor Insights API charges may differ.
