> ## Documentation Index > Fetch the complete documentation index at: https://docs.adsmcp.si/llms.txt > Use this file to discover all available pages before exploring further. # get_chatgpt_insights > Raw ChatGPT Ads delivery insights (impressions, clicks, spend, CTR, CPC, CPM) at account, campaign, ad group or ad level. > [!NOTE] > > **Read-only.** Runs right away and changes nothing in your ad accounts. It still needs a connected platform (check `get_connections_status`) and counts as one tool call. Raw ChatGPT Ads delivery insights (impressions, clicks, spend, CTR, CPC, CPM) at account, campaign, ad group or ad level. Conversions are not available here; use get\_chatgpt\_conversion\_insights. **Platform:** ChatGPT Ads ยท **Type:** Read ## How to call it * **MCP:** on `https://mcp.adsmcp.si/mcp`, run it through the `chatgpt_ads_read` tool with `{"action": "execute", "tool_name": "get_chatgpt_insights", "arguments": { ... }}`, or call `get_chatgpt_insights` by name. `search_tools` finds it and `get_tool_schema` returns its arguments; `https://mcp.adsmcp.si/mcp/flat` lists every tool by name. * **REST:** `POST https://mcp.adsmcp.si/api/v1/tools/get_chatgpt_insights/execute` with `Authorization: Bearer sk_live_...` and a JSON body `{"platform": "chatgpt", "arguments": { ... }}`. The arguments are listed below; they are exactly the tool's input schema. ## OpenAPI ````yaml api-reference/openapi.json POST /api/v1/tools/get_chatgpt_insights/execute openapi: 3.1.0 info: title: ADSMCP REST API version: 0.1.0 description: >- ADSMCP REST API: every public ADSMCP tool as POST /api/v1/tools/{name}/execute. Generated from the ADSMCP code (the service serves the same document at GET /api/v1/openapi.json). Changes to ad accounts are staged for approval by a person on the ADSMCP dashboard; nothing in this API can approve them. servers: - url: https://mcp.adsmcp.si description: ADSMCP MCP and REST service security: [] paths: /api/v1/tools/get_chatgpt_insights/execute: post: tags: - chatgpt summary: >- Raw ChatGPT Ads delivery insights (impressions, clicks, spend, CTR, CPC, CPM) at account, campaign, ad group or ad level. Conversions are not available here; use get_chatgpt_conversion_insights. description: >- Raw ChatGPT Ads delivery insights (impressions, clicks, spend, CTR, CPC, CPM) at account, campaign, ad group or ad level. Conversions are not available here; use get_chatgpt_conversion_insights. Use this when you need that result and get_connections_status already shows the platform as connected. Read tools run without the approval queue. operationId: execute_get_chatgpt_insights parameters: - name: Authorization in: header required: true schema: type: string description: Bearer token - name: Idempotency-Key in: header required: false schema: type: string requestBody: required: false content: application/json: schema: type: object properties: platform: type: string default: chatgpt arguments: type: object properties: account_id: type: string pattern: ^adacct_[A-Za-z0-9_-]{1,120}$ description: >- ChatGPT Ads account id (adacct_...). Optional when exactly one account is connected; with several, call list_chatgpt_accounts first. since: type: string pattern: ^\d{4}-\d{2}-\d{2}$ description: >- First day, YYYY-MM-DD, in the account time zone. Use with until; do not combine with days. until: type: string pattern: ^\d{4}-\d{2}-\d{2}$ description: >- Last day (inclusive), YYYY-MM-DD, in the account time zone. Must be a complete day (before today). days: type: integer minimum: 1 maximum: 90 description: >- Last N complete days in the account time zone (default 14, max 90). Do not combine with since/until. level: type: string enum: - account - campaign - ad_group - ad description: Row level (default campaign). id: type: string pattern: ^[A-Za-z0-9_-]{1,120}$ description: >- Restrict to one entity of the chosen level (campaign, ad group or ad id). Not for level account. fields: type: array items: type: string pattern: ^[a-z_]+(\.[a-z_]+)?$ maxItems: 40 description: >- Canonical insight field names, e.g. campaign.name, ad.title, impressions, spend. Default: id, name, impressions, clicks, spend, ctr, cpc, cpm. segment: type: string enum: - country - device description: Split rows by country or device. filters: type: array maxItems: 10 items: type: object properties: field: type: string operator: type: string enum: - IN - GREATER_THAN - LESS_THAN value: {} required: - field - operator - value additionalProperties: false description: Row filters, all ANDed. sort: type: array maxItems: 3 items: type: object properties: field: type: string direction: type: string enum: - asc - desc required: - field - direction additionalProperties: false description: Sort order. include_zero_impressions: type: boolean description: >- Include entities with zero impressions. Not allowed together with segment. limit: type: integer minimum: 1 maximum: 500 description: >- Maximum rows to return (default 100). The result says when it was truncated. additionalProperties: false additionalProperties: false responses: '200': description: Tool execution result. content: application/json: schema: $ref: '#/components/schemas/ExecuteToolResponse' '400': description: Invalid JSON input. '401': description: Authentication error. security: - bearerAuth: [] components: schemas: ExecuteToolResponse: type: object properties: ok: type: boolean platform: type: string tool: type: string data: {} error: anyOf: - type: 'null' - $ref: '#/components/schemas/ErrorResponse' mcp_response: {} required: - ok - platform - tool - error - mcp_response additionalProperties: false ErrorResponse: type: object properties: code: {} message: type: string details: {} required: - message additionalProperties: false securitySchemes: bearerAuth: type: http scheme: bearer bearerFormat: API token ```` This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.