Frequently asked questions
Short answers to the questions evaluators ask most. What changes in our environment, how this differs from native tools, how fast, what it costs, and what is out of scope.
The questions below come up in most evaluations. Each answer links to the page with the detail.
Does CloudIQ change anything in our environment?#
No. Read-only is enforced in code before any API call is made, and every change leaves as a pull request, a runbook or a ticket for your engineers to apply. The one exception is optional BigQuery reservation automation, which is off by default and outside the standard deployment. See the security and access model.
How is this different from the tools our cloud providers give us?#
Cost Explorer, Compute Optimizer and Google Cloud Recommender are inputs, not competitors. CloudIQ pulls in their recommendations and adds analysis they lack, such as BigQuery pricing fit and commitment options from your own usage. It then turns each finding into work that is ready for review. CloudIQ and native cloud tools has the detail.
How fast do we see results?#
Setup takes about two hours and is read-only. First priced findings follow within the hour, and the first changes are prepared for review the same day. The first savings reach the verified stage in days, as your team ships.
What savings should we expect?#
Industry analysis puts cloud waste at roughly 30% of spend. Typical reductions are 20 to 40%. CloudIQ prices every opportunity before you act, and reports a saving as verified only after a later audit confirms it. Potential and verified amounts are never mixed and never annualized.
How do we know what to work on first?#
The queue ranks findings across all services by expected financial impact. Each item carries its projected saving, evidence, risk and confidence, so the team starts with the highest-value work and makes informed trade-offs.
What does verified mean?#
A task, meaning a finding someone is working on, is verified only when a later audit confirms the condition is gone and later billing data reflects the change where available. Potential savings are never reported as verified.
What happens if a change causes a problem?#
Every task includes a rollback plan, a risk level and a confidence rating. Terraform changes arrive as pull requests and can usually be reversed with a standard Git revert. Actions that cannot be reversed cleanly state the recovery steps. Teams typically start with low-risk changes in non-production and expand from there.
Is our cost data used to train AI models?#
No. Your data sits in an isolated dataset. AI features run under per-customer scopes and budgets, API keys can be revoked in seconds, and nothing is used for training.
Do you support multiple accounts and projects?#
Yes. On AWS, one read-only role per account, with account-scoped views throughout the product. On Google Cloud, grants at project or folder level, and a folder grant covers projects created later.
Can we start with part of our estate?#
Yes, and most teams do. Scope by account on AWS, or by project or folder on Google Cloud, and add more once you are comfortable. A BigQuery-only or SageMaker-only start is common because it needs the least access.
What is out of scope today?#
Azure, which is on the roadmap. Installed agents, which are not part of the architecture. Customer data-plane content, which the access policy denies.
What does a pilot cost?#
Nothing. A limited pilot is free for selected launch customers, with no fees and no commercial commitment. If the platform delivers value and you choose to continue, commercial terms are discussed then. If not, the pilot ends with no obligation.
Can you show results from other companies?#
Yes, on request. The case studies come from the savings practice behind the product. CloudIQ is new, so no product deployment figures are published without evidence behind them.
Next#
Try this against your own spend
A free savings analysis runs the checks described in these docs against your AWS or Google Cloud account and returns a ranked list of priced findings.