Add up idle instances, unattached volumes, orphaned snapshots, over-provisioning, unused commitments and untagged spend.
Waste is split by the action it needs rather than by service: deleting an unattached volume is a one-line change, while resizing an instance family or restructuring commitments needs testing and coordination. Over-provisioning and untagged spend are discounted by a recoverable percentage because neither is entirely waste — headroom is deliberate and some untagged resources are genuinely in use. Industry surveys consistently put wasted cloud spend near a third of the bill, and quantifying it by category turns a vague efficiency goal into a ranked backlog with a dollar value attached. Pull the category figures from your provider's cost tools rather than trusting the defaults.
Cloud Waste
waste = idle + unattached volumes + orphaned snapshots + over-provisioned × recoverable% + unused commitments + untagged × assumed waste% + non-production × idle hours%.
waste = idle + unattached volumes + orphaned snapshots + over-provisioned × recoverable% + unused commitments + untagged × assumed waste% + non-production × idle hours%. Waste is split by the action it needs rather than by service: deleting an unattached volume is a one-line change, while resizing an instance family or restructuring commitments needs testing and coordination. Over-provisioning and untagged spend are discounted by a recoverable percentage because neither is entirely waste — headroom is deliberate and some untagged resources are genuinely in use.
Industry surveys consistently put wasted cloud spend near a third of the bill, and quantifying it by category turns a vague efficiency goal into a ranked backlog with a dollar value attached. Pull the category figures from your provider's cost tools rather than trusting the defaults.
This calculator takes 11 inputs: Total monthly cloud spend, Spend on idle instances, Spend on unattached volumes, Spend on orphaned snapshots, Spend on over-provisioned instances, Share of that recoverable by resizing, Commitment capacity going unused, Spend with no owner tag, Assumed waste inside untagged spend, Non-production spend, Non-production hours nobody uses it. The pre-filled defaults are a realistic starting point — replace them with figures from your own environment for a result you can act on.
Untagged does not mean unused — plenty of critical infrastructure was built before the tagging policy existed. What untagged reliably means is that nobody is accountable for it, and unowned resources are far more likely to be abandoned, so a fraction is counted as waste and the rest as a governance problem.