Cloud Logging and Monitoring cost control comes down to deciding, deliberately, what to keep and for how long. Cloud Logging charges mostly for the volume of log data ingested and for retention beyond the free allotment, while Cloud Monitoring charges for the volume of metric data and chargeable API calls. Both default to capturing more than most teams need, so the savings sit in filtering noise out before it is stored, not in deleting after the fact.
This how-to is part of our Google Cloud cluster. The wider context lives in our complete guide to Google Cloud cost optimization, the pillar this piece links up to. It is a close cousin of persistent disk and storage cleanup on GCP, because logs are just another store that grows until someone trims it.
Find where the volume is coming from
Before you cut anything, see what you are paying for. Cloud Logging reports ingestion by resource type, log name, and project, so you can rank the loudest sources. The usual offenders are data access audit logs, verbose application logging left at debug, load balancer request logs, and chatty third-party agents. Cloud Monitoring shows billable metric volume by metric type. Spend an hour ranking sources by volume and you will usually find that a handful of log names produce most of the cost, which means a few targeted changes do most of the work.
Use exclusion filters to stop noise at the door
The most effective lever in Cloud Logging is the exclusion filter on the _Default sink. An exclusion drops matching log entries before they are ingested and billed, so it saves the full ingestion cost rather than just retention. Common safe exclusions are health-check requests, successful load balancer 200s, and verbose debug entries in production. Be deliberate: exclude noise, not the security and audit logs you may need for compliance, and document every exclusion so a future incident review knows what is and is not captured.
Right-size retention and route to buckets
Logs retained beyond the free period cost storage every month they sit there. Set retention on each log bucket to match how far back you would realistically investigate, which for most operational logs is far shorter than the default people leave in place. Route high-value, long-retention logs such as audit trails to a dedicated bucket with longer retention, and let routine operational logs expire quickly in another. For very long archival needs, exporting selected logs to a cheaper Cloud Storage class can beat paying logging retention; see Google Cloud storage classes and lifecycle management.
Want your observability bill cut without losing signal?
Our Google Cloud cost audit ranks your logging and monitoring volume, writes the exclusion filters, and tunes retention so you keep what matters and stop paying for noise. On the performance model, you pay only from realized savings. No savings, no fee.
Book a GCP cost audit →Control metric volume in Cloud Monitoring
Monitoring costs scale with the number of distinct time series you create, and the silent driver is label cardinality. A custom metric with a high-cardinality label such as user ID or request ID can explode into millions of series and a surprising bill. Keep labels low-cardinality, drop custom metrics no dashboard or alert actually reads, and lengthen sampling intervals where second-level resolution is not needed. Most monitoring overspend is metrics nobody looks at, generated out of habit rather than need.
Set a budget and watch the trend
Logging and monitoring spend grows with the estate, so a one-time cut decays. Put a budget on the observability spend, alert when it trends up, and review the loudest sources quarterly as services ship and change. Wire it into your wider controls with budgets and alerts in Google Cloud. This is the Lock and Run half of the method: the cut holds only if something watches it.
| Lever | Where | Saves |
|---|---|---|
| Exclusion filters | Cloud Logging sink | Ingestion cost |
| Retention tuning | Log buckets | Storage cost |
| Archive to GCS | Cloud Storage | Long-term retention |
| Low-cardinality labels | Cloud Monitoring | Metric volume |
| Budget and review | Billing | Holds the savings |
Product names and pricing mechanics above reflect Google Cloud as of May 2026. Verify current logging and monitoring pricing and free allotments in Google Cloud documentation before acting, as they change.
The Google Cloud Cost Optimization Field Guide includes the exclusion filter patterns and retention model behind this article. It is the downloadable companion.
The short version
Rank your log and metric volume, use exclusion filters to stop noise before ingestion, set retention to match real investigation windows, archive long-term logs to cheaper storage, keep metric label cardinality low, and put a budget on the trend. Pair this with storage cleanup on GCP to clear the rest of the silent spend. When you want the whole observability bill audited and cut for you, that is what our Google Cloud cost optimization service delivers.