Workspace

Usage and budgets

See every AI call and its cost, and set a monthly budget that pauses batches.

Settings, Usage shows what your AI key spent through Echo. Costs are calculated from the exact tokens your provider reports, at your provider's price, with no markup.

Settings, Usage: spend, projection, cost per lead and tokens.

What you see

  • Spend to date, Projected month, Per lead (AI slots) and Tokens (with how much came from cache).
  • Daily spend and spend per provider.
  • Where it goes, grouped by feature (Builder chat, AI slots per lead, Website research), model, campaign or teammate.
  • Export CSV downloads every call.

Set a budget

Budget and alerts: the monthly budget, the alert threshold, where alerts go, and a test button.

Under Budget and alerts, set a Monthly budget (USD) and an Alert at percentage, and choose where alerts go: Slack (your connected channel), Email (workspace admins), or both. Echo sends one alert when spend passes the alert percentage and another at 100%, at most once each per month. Send test alert checks the delivery now.

Turn on Pause AI slot batches when the budget is reached to stop campaign builds at the limit; the builder keeps working. Raise the budget to continue.

Something unclear or missing? Email support@relevanceai.io and we will fix the page.