Audit your team's Claude spend — free, in 60 seconds
Paste your Anthropic Spend Report CSV. Eight checks run right in your browser — over-tiered and idle seats, who's driving the bill, your model mix, and more. Nothing is uploaded; your data never leaves your device.
A Claude Team spend audit answers one question: where is the money going, and what could you right-size? This free tool reads your Anthropic Spend Report (the per-user, per-model CSV you export from Settings → Analytics) and runs eight checks on it — entirely in your browser. It finds over-tiered and dormant seats, spend concentration, model-mix waste, hidden metered surfaces, and $/request outliers, and hands you recommendations you confirm. It never uploads your file and never changes anything in your account.
Your Anthropic Spend Report (claude.ai → Settings → Analytics → Export Spend Report; Owner access, last 90 days).
Parsing and all 8 checks run locally, instantly. Nothing is uploaded.
Right-size over-tiered seats, reclaim dormant ones, trim model mix. You confirm every change.
Get your Spend Report (≈30 seconds)
- 1. claude.ai → your initials → Settings
- 2. Analytics → Spend → Export Spend Report
- 3. Choose Last 90 Days → download the CSV
You must be an Owner / Primary-Owner (not the Team “Admin” role). We ingest this native export — we never re-create it. This is the per-person Spend Report, not the Console / API billing export from platform.claude.com.
Your spend data never leaves your browser. The whole audit runs on your device — nothing is uploaded, nothing is stored.
Drop your Spend Report CSV here
or click to choose a file — it never leaves your browser
Optional — unlocks idle-seat reclaim $ (else we report dead-weight by usage volume only).
Paste CSV instead
FAQ
Is this Claude spend audit really free? +
Yes — the full 8-check audit is free and runs in your browser. The optional continuous-monitoring product is separate and not yet available (you can join the waitlist).
Do you upload or store my spend file? +
No. Everything runs client-side; your CSV never leaves your device.
What do the 8 checks cover? +
Dead-weight seats, spend concentration, model mix, hidden metered surfaces, $/request outliers, discount exposure, domain allocation, and context ratio.
Which file do I need? +
The per-user Anthropic Spend Report from claude.ai → Settings → Analytics (not the API/Console billing export). If you upload the wrong one, the tool tells you and points you to the right export.
Will it change anything in my Claude account? +
Never. Every output is a recommendation you review and apply yourself.
See everything WTClaude does.
Billing-grade accuracy, backup & multi-machine sync, a full dashboard, billing-change readiness — the most complete Claude Code tracker.
Explore all features →