Practical guide

CloudWatch Logs pricing: estimate ingestion from log volume

estimate CloudWatch Logs ingestion cost from application traffic. Includes a worked example, calculator scope, and practical questions.

Updated · Sources checked

Estimate uncompressed ingestion first

CloudWatch Logs ingestion is a distinct charge from archive storage and analysis. A useful starting point is the amount of uncompressed log data the application sends, using the correct log class and source category. Compressed archive size is not an interchangeable ingestion input.

The StackCalcs tool focuses on first-tier Standard ingestion in US East. It does not represent a complete observability bill, and it does not apply the shared free tier. Use it when the question is specifically how a change in ingested GB affects that component.

Worked example: requests become log bytes

Assume an application receives 12 million requests in a month and writes three log events per request. If measured average serialized event size is 600 bytes, request logs contribute 12,000,000 × 3 × 600 = 21.6 billion bytes.

Add a hypothetical 5.4 billion bytes from background jobs and startup events. The total is 27 billion bytes, or 27 decimal GB for the example. Convert that byte total into the unit used by the billing input before estimating cost.

Now consider reducing successful-request logging from three events to one, without changing the 600-byte average. Request logs become 7.2 billion bytes; with the same background logs, the total is 12.6 billion bytes. The ingestion reduction is 14.4 billion bytes. This is an original workload scenario, not an assumed compression ratio or provider discount.

Validate before changing production logging

Sample a representative normal day and a failure-heavy day. Stack traces can make error events much larger than routine messages. Retries and deployment startup logs can also break a forecast based only on business requests.

Retain the diagnostic and audit events the application needs. Removing repetitive success messages is different from losing evidence needed to investigate errors. Compare measured ingestion after the change rather than assuming event count and byte count fall by the same percentage.

Frequently asked questions

Does shorter retention reduce ingestion charges? It changes retained storage, not the bytes already ingested. Model the two components separately.

Can I enter the compressed archive size? No. Use ingestion volume for the relevant billing category; archive storage and compression have different roles.

Calculate your scenario

Use the cloudwatch logs cost calculator. Keep the displayed region, pricing date, billing units, and exclusions alongside your result. The arithmetic examples above illustrate usage or explicitly hypothetical rates; they are not AWS quotes.

Source

Billing structure checked against official AWS documentation on September 6, 2026.

Official sources

Related calculators

Continue reading