A Google Cloud cost optimization checklist is only useful if it is ordered by value and risk. The 30 quick wins below follow our See, Cut, Lock, Run method: see the spend clearly, cut waste and rate, lock the savings with governance, and run it continuously. Work the list roughly in order, because rightsizing before committing, and cleaning waste before discounting, is what makes each later step pay more.
This checklist is part of our Google Cloud cluster. For the deeper treatment of any item, start with our complete guide to Google Cloud cost optimization, the pillar this piece links up to.
See: get the spend visible (wins 1 to 6)
1. Open Cloud Billing reports and build views by project, service, and label. 2. Enforce a label standard so every resource carries a team, environment, and cost-center value; see Cloud Billing reports and BigQuery billing export. 3. Turn on detailed billing export to BigQuery so you can query spend at SKU granularity. 4. Organize projects under folders that mirror how you want to allocate cost. 5. Identify your top ten cost drivers; on most estates they account for the bulk of the bill. 6. Establish a baseline run rate before you change anything, so every saving is measurable.
Cut compute (wins 7 to 13)
7. Apply Compute Engine rightsizing recommendations from Active Assist; see how to rightsize Compute Engine VMs with Recommender. 8. Move to current-generation machine families for better price-performance. 9. Stop or suspend non-production VMs outside working hours with instance schedules. 10. Move fault-tolerant and batch work to Spot VMs for steep discounts off on-demand. 11. Delete idle VMs flagged by the idle VM recommender. 12. Rightsize and bin-pack GKE with Autopilot or node pool autoscaling; see GKE cost optimization. 13. Move spiky request-driven work to Cloud Run and Cloud Functions so it scales to zero.
Cut storage and data (wins 14 to 20)
14. Apply Cloud Storage lifecycle rules to move cold objects to Nearline, Coldline, and Archive. 15. Delete orphaned persistent disks left behind by removed VMs. 16. Clean up old disk snapshots and redundant images. 17. Right-size Cloud SQL instances and use the right edition; see Cloud SQL cost optimization. 18. Cap BigQuery on-demand query spend or move heavy workloads to capacity; see BigQuery on-demand vs editions. 19. Partition and cluster large BigQuery tables and set table expiration on staging data. 20. Review long-term storage pricing; BigQuery storage drops in price after 90 days untouched, so prune what you do not query.
Cut the rate (wins 21 to 25)
21. Let sustained use discounts apply automatically on eligible long-running VMs; see sustained use discounts on Google Cloud. 22. Buy committed use discounts on the steady baseline after rightsizing; see committed use discounts explained. 23. Use spend-based (flexible) CUDs where machine families shift, resource-based CUDs where they are stable. 24. Negotiate an enterprise discount on the optimized run rate; see how to negotiate a Google Cloud commitment. 25. Review CUD utilization monthly so you never carry unused commitment.
Want the whole checklist run for you?
Our Google Cloud cost audit works every item above across your estate, ranks each by dollars, and hands you a prioritized plan. On the performance model, you pay only from realized savings. No savings, no fee.
Book a GCP cost audit →Cut networking and the long tail (wins 26 to 28)
26. Review network egress and inter-region traffic; keep chatty services co-located in one region. 27. Delete idle load balancers, unused external IP addresses, and orphaned NAT gateways. 28. Trim Cloud Logging and Monitoring ingestion with exclusion filters and retention limits.
Lock and Run: keep it cut (wins 29 to 30)
29. Set budgets and threshold alerts per project and team so spend drift is caught early. 30. Add Organization Policy guardrails and schedule a monthly cost review so rightsizing and commitments stay fresh. These last two are the difference between a one-time cut and a unit cost that keeps falling.
| Phase | Wins | Typical effort |
|---|---|---|
| See | 1 to 6 | Low, do first |
| Cut compute, storage, rate | 7 to 25 | Where most savings sit |
| Cut networking and tail | 26 to 28 | Smaller, worth it |
| Lock and Run | 29 to 30 | Keeps savings durable |
Features, tiers, and program names above reflect Google Cloud as of May 2026. Verify current pricing and options in Google Cloud documentation before acting, as the platform changes.
The Google Cloud Cost Optimization Field Guide turns this checklist into a worked playbook with the queries and scoring model behind each item. It is the downloadable companion to this article.
The short version
Make spend visible, cut compute and storage waste, apply sustained and committed use discounts on the clean baseline, trim the networking tail, and lock it all in with budgets, policy, and a monthly review. The biggest single rate item for most Google Cloud estates is the commitment, so once the baseline is clean, read how to negotiate a Google Cloud commitment. When you want the whole list run across the estate, that is exactly what our Google Cloud cost optimization service delivers.