Savings playbook

A prioritised, step-by-step programme for reducing your cloud bill, with a 30/60/90-day plan.

Savings playbook: how to reduce your cloud bill

A practical order of work for cutting cloud costs: remove waste first, then right-size, then commit, then improve the architecture. Each step names the CostGlass page that supports it.

Where in CostGlassCosts → Savings

PlanAll plans (Commitments and Instance Scheduler need Growth or above)

Who can do thisEngineering, finance and an executive sponsor (see Ownership)

Time neededA 90-day programme

In short

Work in this order: 1. remove waste, 2. schedule non-production, 3. right-size, 4. commit, 5. improve the design. Each step makes the next one safer.
Commit after you right-size. Otherwise you lock in a bill that was about to fall.
Do not add up the savings cards. CostGlass counts overlaps once, and so should you.
Prove each saving on your real bill. A saving you cannot see in Cost Explorer next month has not happened.
Give every step a named owner and a date.

Four principles

Safest first: Start with changes that are easy to reverse and happen away from production: deleting things nobody uses, switching off test servers at night.
Measure before you change: Use real utilisation and real billed cost. CostGlass tags every figure with how it was reached (Measured, List price, Cloud estimate or Estimate), so you know which ones to confirm.
Commit last: A commitment is a promise you cannot cancel. Make the bill as small and as steady as possible before you promise to pay it.
Verify on the bill: CostGlass records what you mark as done, but only your invoice proves the saving.

Note

CostGlass is read-only. It finds opportunities and gives you the steps; your team makes the changes in the cloud console. The one exception is the opt-in Instance Scheduler.

Before you start

Before you begin

All your clouds connected and showing data. See Connect your clouds.
Resource tags that say who owns what. See Tagging strategy; without them you will not know whom to ask before deleting something.
A baseline: note your last full month’s total in Cost Explorer. You will compare with it.
A sponsor who can approve small risks and a short list of engineers with time to act.

Run a full scan

Open Savings and click Run Full Scan. Read What is checked for each cloud & platform to learn what is covered and what is not.

Check for missing permissions

If a check says it found nothing because access is missing, apply the one-off update described in Savings recommendations.

Note the headline

Write down Potential Savings and today’s date. This is your starting list, not a promise.

The programme, step by step

Step 1: Remove waste (quick wins)

Waste is anything you pay for and do not use. It is the cheapest saving because nothing needs to be redesigned.

What to look for	Where in CostGlass	Safety check
Unattached disks and volumes, unassigned IP addresses, powered-off servers that still bill, old snapshots	Savings (filter by Low Effort)	Ask the owner. Take a snapshot or backup first if unsure. Snapshot figures are an upper bound.
Idle load balancers and idle servers or databases	Savings, Rightsizing (Idle)	Idle means measured low use over 12.5 days or more. Check for monthly or seasonal jobs.
A service whose cost jumped and stayed up	Savings (sustained cost increase), Anomalies	It may be a planned launch. Ask what changed.
Resources nobody owns	Tag compliance	Find the owner before acting.

Warning

Deleting is permanent. Never delete only because a recommendation says "unused". Confirm with the owner.

Step 2: Schedule non-production

Development, test and staging servers rarely need to run at night or at weekends. Use the Instance Scheduler to stop them outside working hours (a Growth plan or above, and a one-off permission). Start with a few servers, watch for problems for a week, then roll it out. Do not schedule databases or stateful servers without care.

Step 3: Right-size

Move over-sized servers and databases down one size at a time, using measured CPU and memory. See Rightsizing. It needs about a week of metrics (12.5 days for idle checks), and memory data needs the CloudWatch agent on AWS. Test on non-production, change in a maintenance window, and keep the old size as your rollback.

Step 4: Commit

Once the bill has settled, cover your steady baseline with Savings Plans, Reservations or committed use discounts. Use Commitments to see coverage, utilisation, recommendations and expiry. Aim for the 70% to 85% band, prefer a one-year term at first, and never commit to your peak.

Warning

If you commit before steps 1 to 3, you may promise to pay for servers you are about to shrink or delete.

Step 5: Improve the architecture and AI choices

These take the longest and carry the most design risk, so they come last. Examples CostGlass can point to: previous-generation instance types and Graviton or ARM moves (the extra opportunities under Rightsizing), Kubernetes workloads that request far more than they use (Kubernetes costs), and for AI platforms, moving non-urgent work to the Batch API, using prompt caching and choosing a cheaper model tier where quality allows (SaaS and AI spend). The AI savings are labelled estimates, based on stated assumptions. Test quality before switching a model.

A 30/60/90-day plan

Period	Goal	Actions	Done when
Days 1 to 30	Visibility and quick wins	Connect everything. Run a full scan. Fix tags. Clear the low-effort waste. Pilot the Instance Scheduler on a few non-production servers. Set a budget and cost alerts.	Waste items are marked done or dismissed with a reason. A baseline is recorded.
Days 31 to 60	Right-size and schedule widely	Roll scheduling out. Work through High-confidence downsize rows, starting in non-production. Install the CloudWatch agent where memory is unknown. Review the Kubernetes and AI items.	Rightsizing savings appear on the bill. No performance incidents from resizing.
Days 61 to 90	Commit and institutionalise	With a settled bill, buy commitments to a safe coverage level. Set renewal reminders. Add a monthly savings review.	Coverage is in the target band with healthy utilisation. A repeatable monthly review exists.

Tip

Use Budgets and Cost alerts in the first month so new waste is caught as it appears, not at the next review.

How to measure progress

Realized Savings: On the Savings page, the monthly saving of items you marked Mark done. It records your claim; it does not verify it.
The bill itself: In Cost Explorer, compare the month after a change with the month before, for the same service and account. Consider usage growth, so look at cost per unit if you can (Unit economics).
Commitment health: Coverage inside 70% to 85% and AWS utilisation at or above 95%.
Waste trend: The number and value of Open recommendations falling over time, and not returning after rescans.

For example

Suppose you marked a recommendation done that promised 200 a month. Next month Cost Explorer shows that service fell by only 120, but your traffic also grew. Record the saving you can defend (120), investigate the rest, and carry on. (The figures are only an illustration.)

Avoid counting the same saving twice

A Savings Plan, a rightsizing change and a schedule can all reduce the same servers. If you add the cards, the total can exceed what you spend. CostGlass handles this for you: Potential Savings applies recommendations one after another to each cloud’s bill, so overlapping savings are counted once and the total can never go above the bill. The individual cards will therefore sum to more than the headline. Quote the headline, and keep the order of work in this playbook so each step’s saving is real before you rely on the next. See Savings recommendations for the detail.

Note

Percentage-based estimates (for example commitment savings where your cloud has not yet supplied a recommendation) use planning rates stated on the card. They are not quotes.

Who should own what

Role	Responsibilities
Executive sponsor	Sets the target, unblocks risk decisions, reviews progress monthly.
FinOps lead or finance partner	Owns the baseline, the monthly review, commitment decisions (with engineering input) and the budget.
Engineering owners (per team or service)	Review and carry out waste, scheduling and rightsizing changes for their own resources. They know what is safe.
Platform or DevOps team	Install the CloudWatch agent, deploy the scheduler permission, run Kubernetes and architecture changes.
Cloud administrators	Make purchases in the cloud console and keep stacks and permissions up to date.

Use Team access and Organisation hierarchy so each team sees only its own recommendations.

Frequently asked questions

Where do I start if I only have an hour?

Run a full scan, filter by Low Effort, and clear unattached disks and idle resources in development accounts.

Why not buy a Savings Plan straight away?

Because you would commit to a bill that waste removal and right-sizing are about to reduce. Commit last.

How much will we save?

It depends on your estate. The Savings page shows figures for your own data, each labelled with how it was reached. We do not promise a percentage.

What if a change causes a problem?

Make changes in small steps, in non-production first, in a maintenance window, and keep the previous setting so you can roll back.

How often should we review?

Monthly for the full list, with cost alerts and budgets catching new surprises in between.

All Documentation Guides

Welcome to CostGlass — What CostGlass does, who it is for, how your data gets in, the plans and regions, and a plain-English glossary.
Quickstart: your first hour — Sign up, connect a first cloud, invite your team, set a budget and turn on anomaly alerts, with a first-week checklist.
Finding your way around — A tour of the sidebar, top bar and page controls: what every page answers, plan gating, currency, language, links and caching.
Dashboards — The Overview dashboard widget by widget, plus building custom dashboards you can share with your organisation.
Cloud Resources — The multi-cloud inventory: filters, search, categories, estimated cost, terminated resources and CSV or Excel export.
Uptime, incidents & support requests — Monitor public endpoints, route alerts, see incidents and raise requests to CostGlass engineers.
How CostGlass accesses your data — Security and data ingestion architecture: how CostGlass reads multi-cloud billing and telemetry through read-only roles and least-privilege permissions.
Dashboard & Platform Screenshot Tour — Visual walkthrough of the CostGlass interface with screenshots covering multi-cloud spend, Cost Explorer, anomalies, Kubernetes allocation, and savings.
Connecting your clouds: overview — Every cloud and service you can connect, how each one connects, what access it needs, how long data takes and your plan limits.
Connect AWS (CloudFormation) — Deploy the generated CloudFormation template, choose organization or single account and optional features, then verify.
Connect AWS with Terraform instead of CloudFormation — A Terraform equivalent of the generated AWS template, with an Organization variant and a way to compare it to the wizard.
Connect Azure — Admin consent for the CostGlass application, Reader and Cost Management Reader roles, management group or subscription scope, and a Terraform version.
Connect Google Cloud — Enable the FOCUS BigQuery export, run the IAM grants CostGlass prints, verify, and a Terraform version.
Managing connected accounts — The Cloud Accounts page: statuses, syncing, errors, naming, disconnecting, and the FOCUS 1.0 and CUR Export pages.
Connect Oracle Cloud (OCI) — Give CostGlass read-only access to Oracle Cost and Usage Reports with a Resource Manager stack or your own Terraform.
Connect Alibaba Cloud — Connect Alibaba Cloud through Bill Subscription files in OSS and a read-only RAM role created by a ROS template.
Connect DigitalOcean — Connect DigitalOcean accounts with a read-only personal access token and see invoices by month.
Connect Akamai / Linode — Connect Linode (Akamai Connected Cloud) accounts with an account read-only token.
Connect Cloudflare — Connect Cloudflare usage-based charges with a Billing Read account token and see credits and unpaid debt.
Connect OpenAI, Anthropic, Datadog and Snowflake — Connect AI and software vendors with admin keys or a Snowflake key pair so their spend sits beside your cloud costs.
Cost Explorer — Slice, compare and drill into spend across every connected cloud: group by, filters, charts, overlays, details, AI, links and export.
Saved views, saved filters & scheduled emails — Keep, share and email the Cost Explorer views you use often, and reuse include/exclude filters across views and reports.
Reports — Build, compare, export, save and schedule spend reports, with presets, Excel, CSV and PDF exports.
SaaS & AI Spend — Spend and token analytics for OpenAI, Anthropic, DeepSeek, Datadog, Snowflake and cloud AI services.
Unit Economics — Cost per business unit: define metrics, send volumes, set targets and alerts.
Connect a Kubernetes cluster — Install OpenCost and the CostGlass exporter in your cluster so cost data is pushed to CostGlass every 15 minutes.
Kubernetes Costs page — Understand cluster, namespace, workload and pod costs, idle capacity, rightsizing, carbon and health.
Tagging strategy: getting the most from tags — How to tag cloud resources, enforce tagging, and when to use cloud tags or virtual tags in CostGlass.
Tag Explorer — Group billed spend by the value of a tag, see tag coverage per cloud and drill into resources.
Tag Compliance — Define required AWS tags, measure compliance, and fix missing tags with Quick Tag, batch apply or a CLI script.
Virtual tagging — Allocate spend by team, product or environment with rules, no cloud changes, with a rule builder and CostGlass AI.
Forecasting — Forecast month-end spend, the next months and a yearly run-rate, and simulate the cost of changes.
Budgets — Create spending limits for a cloud, account, service or team, and see how CostGlass budgets differ from native AWS, Azure and Google budgets.
Anomaly detection — How CostGlass finds unusual daily cost, how to tune it, explain it and be notified.
Cost alerts — Alert when spend rises versus the previous day, week, month or quarter.
All alert and notification types — A comparison of every alert and notification in CostGlass, with channel setup and testing.
Issues & annotations — Dated notes on spend, and a lightweight issue tracker for cost work with Jira push.
Savings playbook — A prioritised, step-by-step programme for reducing your cloud bill, with a 30/60/90-day plan.
Savings recommendations — How the Savings page finds, prices and tracks savings for every connected cloud and platform.
Rightsizing — Find servers and databases that are too big, or too small, using real CPU and memory measurements.
Commitments: Savings Plans & Reserved Instances — Track coverage, utilisation, expiry and purchase recommendations for Savings Plans, Reservations and committed use discounts.
Instance Scheduler — Start and stop non-production servers on a weekly schedule to stop paying for idle hours.
Giving your team access — Invite people, choose Owner, Admin or Member roles, remove users and limit what they see.
Setting up your organisation hierarchy — Build a tree of nodes, place cloud accounts by hand, rule or CSV, and scope people to a branch.
Designing your hierarchy (with enterprise examples) — Principles and worked patterns for AWS Organizations, single accounts, multi-cloud groups, MSPs and regulated regions.
Single sign-on (SAML & OIDC) — Connect Okta, Entra ID, Google Workspace or another provider, verify your domain, test and require SSO.
API keys & AI assistants (MCP) — Create API keys and connect Claude Desktop, Cursor or Claude Code to your cost data.
Audit log — See who changed what and when, with before and after values, kept for 24 months.
Data residency & retention — Your fixed region, how long cost history is kept, purge now and the data handling summary.
Notification settings, Jira & chat — Weekly spend summary, Slack and Teams webhooks, and connecting Jira Cloud.
Troubleshooting & Common Setup Fixes — Solutions for common cloud connection and ingestion issues across AWS, Azure, Google Cloud, OCI, Alibaba Cloud, Kubernetes, and SSO.

Savings playbook