> For the complete documentation index, see [llms.txt](https://docs.finout.io/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.finout.io/billing-integrations/ai-providers/connect-to-anthropic/tracking-claude.ai-spend-teams-plan.md). # Tracking Claude.ai Spend (Teams plan) {% hint style="warning" %} **This page is for Team plans only.** If you are on an Anthropic Enterprise plan, connect Claude.ai natively from the [Connect to Anthropic](https://docs.finout.io/billing-integrations/ai-providers/connect-to-anthropic) wizard instead — no manual workaround needed. Use this page only if you are on a Team plan. {% endhint %} #### Overview This page describes a manual workaround for bringing Claude.ai spend into Finout on a **Team plan**. The native [Anthropic integration](https://docs.finout.io/billing-integrations/ai-providers/connect-to-anthropic) connects both Anthropic surfaces directly — the API platform (Claude Platform) and Claude.ai — but the Claude.ai surface requires an Anthropic Enterprise plan. On Team plans, Anthropic does not expose Claude.ai data through the Analytics API, so the workaround on this page applies instead. The workaround loads a CSV export from the Claude.ai admin console as a [Custom Cost Center](https://claude.ai/custom-cost-sources/custom-cost-centers) in Finout, so Claude.ai spend appears alongside your API platform spend in MegaBill. #### Before you start The Claude.ai spend export only captures spend that exceeds your included usage allotment. If extra usage isn't enabled on your Team plan, no spend report is generated. Flat per-seat fees are billed by Anthropic outside of the spend export and appear only on your Anthropic invoice; they are not covered by the workaround on this page. **Who can run the export** On Team plans, only Owner and Primary Owner roles can view usage analytics and run the spend export. Admins on Team plans cannot. The automation described below needs to authenticate as a role with access. #### Set up a Custom Cost Center for Claude.ai spend The pipeline has three parts: 1. Pull the spend report from Claude.ai on a daily schedule. 2. Transform it to the Custom Cost Center schema. 3. Drop the transformed file into an S3 bucket Finout has access to. The [Custom Cost Center](https://docs.finout.io/billing-integrations/custom-cost-sources/custom-cost-centers) doc covers schema requirements, S3 setup, and ingestion behavior. The sections below cover what is specific to the Claude.ai source. **1. Pull the spend report from Claude.ai** Anthropic does not expose this data through a public API for Team plans, so the daily pull is implemented as a browser-driven automation against the Claude.ai admin console. The manual UI flow looks like this: 1. **Navigate** to **Admin settings > All activity** in Claude.ai. 2. **Scroll** to the **Spend** section. 3. **Click** **Export Spend Report**. 4. **Choose** **Custom** as the time period and set the start and end date to the previous day. 5. **Download** the CSV. It is recommended to replicate this flow in a headless authenticated session running on a daily schedule, using a browser automation framework such as Playwright, Selenium, or an RPA platform. Configuring the export with a one-day window keeps the data daily-granular in Finout. {% hint style="info" %} **The export has a one-day lag.** Today's spend is available tomorrow. Plan dashboards and anomaly detection accordingly. {% endhint %} **2. Transform to the Custom Cost Center schema** The Claude.ai export is structured around per-user, per-model rows summed across the selected date range. To load it into a Custom Cost Center, transform each row to include the two mandatory columns plus any dimensions you want to slice on in Finout. | Custom Cost Center column | Source from Claude.ai export | Notes | | ------------------------- | ---------------------------- | ----------------------------------------------------------------------- | | `UsageDate` | Date of usage | Format `YYYY-MM-DD`. Run a daily export window to get one date per row. | | `Cost` | Net cost in USD | Use the post-discount cost column to match what is actually invoiced. | | `User` | User email | Optional dimension. Enables per-user views and Virtual Tag mapping. | | `Model` | Model | Optional dimension. Sonnet, Opus, Haiku, etc. | **3. Drop the file into an S3 bucket Finout has access to** Provision an S3 bucket on your side and grant Finout access to it, then write the transformed CSV there. Once files are available in the bucket, contact to complete the integration. The full setup steps — bucket configuration, access permissions, and ingestion behavior — are covered in the [Custom Cost Center](https://claude.ai/custom-cost-sources/custom-cost-centers) doc. #### FAQs **Why isn't my Claude.ai spend included in the main Anthropic integration?** On Team plans, Anthropic publishes Claude.ai data separately and does not expose it through the Analytics API. Enterprise customers can connect Claude.ai natively from the [Connect to Anthropic](https://docs.finout.io/billing-integrations/ai-providers/connect-to-anthropic) wizard. Team customers, and those needing pre-2026 historical data, can use the workaround on this page. **The export only shows partial spend on my Team plan. How do I track the rest?** The spend export only contains usage that exceeds your included allotment. Flat per-seat fees are billed separately by Anthropic and aren't part of the export. For help designing the daily extraction, the transformation, or the S3 drop pattern, contact . --- # Agent Instructions This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com. ## Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question. Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter: ``` GET https://docs.finout.io/billing-integrations/ai-providers/connect-to-anthropic/tracking-claude.ai-spend-teams-plan.md?ask=&goal= ``` `ask` is the immediate question: it should be specific, self-contained, and written in natural language. `goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.