A Google Cloud cost assessment is a structured pass over the estate that answers one question: where is the money going, and which moves recover the most for the least risk. Done well, it ends not in a dashboard but in a prioritized plan. The method below follows our See, Cut, Lock, Run framework and produces findings you can act on the same week.
This how-to sits in our Google Cloud cluster. The wider context lives in our complete guide to Google Cloud cost optimization, the pillar this piece links up to, and it pairs with our 30-win Google Cloud checklist for the action items.
Step 1: scope the estate and the baseline
Start by drawing the boundary. List every billing account, the folders and projects beneath them, and the regions in use. Pull the last three to twelve months of spend so seasonality is visible, and lock a baseline monthly run rate. Without a baseline, no saving is provable later. Note which projects belong to which team, because allocation is the spine of every later finding.
Step 2: get the data with billing export
The Cloud Billing reports in the console are enough for a first read, but a real assessment runs on the detailed billing export to BigQuery, which gives you SKU-level cost, usage, credits, and labels. Enable it, wait for a full day of data, then query spend by service, project, label, and SKU. Our guide to Cloud Billing reports and BigQuery billing export covers the setup and the first queries to run.
Step 3: find the waste (the Cut findings)
Waste is the fastest money because it needs no commitment and carries little risk. Work the recommenders and the obvious idle resources. Compute Engine rightsizing and idle VM recommendations from Active Assist surface oversized and unused instances; our walkthrough on rightsizing Compute Engine VMs with Recommender shows how to read them. Then sweep for orphaned persistent disks, old snapshots, idle load balancers, unused external IPs, and over-provisioned Cloud SQL instances. For container estates, check GKE node utilization. Each finding gets a dollar value and an owner.
Step 4: find the rate opportunities
Once you know the rightsized baseline, look at the rate. Sustained use discounts apply automatically, but committed use discounts are where the deliberate savings sit. Map the steady-state compute that has run for months and is unlikely to change, and size a committed use discount against it; see committed use discounts explained. Do not size the commitment against today's oversized footprint, or you lock in waste. For large estates, an enterprise agreement may add a further layer; see how to negotiate a Google Cloud commitment.
Step 5: rank everything by dollars and risk
Every finding now goes into one table: the action, the estimated monthly saving, the effort, and the risk. Sort by saving-per-unit-of-effort. The top of that list is almost always waste cleanup and rightsizing, which is why they come before any commitment. The result is a plan, not a report, and it is the deliverable that makes the assessment worth running.
Want the assessment run by specialists?
Our Google Cloud cost audit runs every step above across your estate and hands you a ranked savings plan with dollar values and risk levels. On the performance model, you pay only from realized savings. No savings, no fee.
Book a GCP cost audit →Step 6: lock and run
An assessment that is not operationalized decays in a quarter. Set budgets and alerts per project, add Organization Policy guardrails, and schedule the assessment to repeat monthly or quarterly so new waste is caught as teams ship. This is the Lock and Run half of the method, and it is what turns a one-time cut into a unit cost that keeps falling.
| Step | Output | Method phase |
|---|---|---|
| Scope and baseline | Run rate, project map | See |
| Billing export | SKU-level data | See |
| Waste findings | Idle and oversized list | Cut |
| Rate findings | Commitment plan | Cut |
| Ranked plan | Prioritized actions | Cut |
| Lock and run | Budgets, recurring review | Lock · Run |
Tooling and recommender names above reflect Google Cloud as of May 2026. Verify current product names and pricing in Google Cloud documentation before acting, as the platform changes.
The Google Cloud Cost Optimization Field Guide includes the assessment scoring model and the BigQuery queries behind each finding. It is the downloadable companion to this article.
The short version
Scope the estate and lock a baseline, get SKU-level data through billing export, find waste first and rate second, rank every finding by dollars and risk, then lock it in with budgets and a recurring review. The assessment is the front door to every other play; when you would rather have it run for you, that is what our Google Cloud cost optimization service delivers.