Projects

Projects group your AI usage so you can track costs, set budgets, and generate reports per client, team, or purpose. Each project has its own API keys, spend limits, and attribution data.

Projects vs workspaces

Your workspace is the top-level container — it holds your billing plan, members, and provider keys. Projects live inside a workspace and subdivide your usage. Think of a workspace as your organization and projects as the work you do for individual clients or teams.

Agencies see "Client" instead of "Project" — same concept, different label. If you selected "Agency" during onboarding, your dashboard shows "Clients" everywhere. The underlying behavior is identical.

When to create a project

Separate client billing — one project per client, so you can report (or bill back) per-client AI spend
Department boundaries — marketing vs. support vs. product, each with their own budget
Environment separation — production vs. staging, so test runs don't pollute cost reports
Workflow grouping — when you have 20+ workflows and want to see costs by business function

When NOT to create a project

If you're a solo builder running 5–15 personal workflows, workspace-level tracking is enough. You don't need a project for every workflow — the attribution system tracks individual workflows automatically. Projects are for when you need separate budgets, separate API keys, or separate cost reports.

Setting up your first project

Create a project from the dashboard — give it a name, optionally set a budget, then create an API key scoped to that project. Any request made with that key is automatically attributed to the project. See Get Started for the full setup flow.

Budgets — set spend limits per project
Attribution — how costs are tagged to workflows within a project
n8n usage — project tagging in n8n workflows