Overview

Jump to AboutAbout

The Enterprise API gives you access to Prompt Tracking data from your AI Optimization (AIO) projects.

Use it to retrieve:

• The prompts and topics tracked in an AIO project.
• Brand visibility, mentions, sentiment, and citations.
• Competitors, cited sources, and AI responses.

Jump to Get accessGet access

The Enterprise API is available for Semrush Enterprise subscriptions. It currently doesn’t consume API units.

Jump to AuthorizationAuthorization

The Enterprise API requires a Version 4 API key, which is assigned to each user after subscribing to Semrush.

You’ll find your API key in the API keys section of your Semrush profile. If your API key is unavailable there, refer to API access.

Pass the key in the Authorization header of every request:

Authorization: Apikey YOUR_API_KEY

For details, refer to Authorization.

Jump to Get startedGet started

1. Call Get Workspaces to get a workspace_id.
2. Pass the workspace_id to Get AIO Projects to get a project_id.
3. Call Get AIO Project and check that aio_data_ready is true. Until then, the Prompt Tracking endpoints return no data.
4. Use both IDs with Get Tracking Dates to find the dates with available project data. Then use the other Prompt Tracking endpoints to retrieve your data.

Requests are subject to rate limits that depend on the endpoint group. For details, refer to Rate limiting.

Jump to FilteringFiltering

Endpoints that document a filter parameter support FilterQL expressions. Use filters to narrow results by one or more filterable fields.

• Combine conditions with AND, OR, and parentheses.
• Supported operators are =, !=, >, >=, <, <=, CONTAINS, HAS, STARTS_WITH, and ENDS_WITH.
• Only fields listed as filterable in the endpoint documentation can be used. Depending on the endpoint, these may include tags, topic, avg_position, and mention_count.
• A filterable field doesn’t need to be included in fields or returned in the response.
• The tags field supports only the HAS operator.
• Enclose strings and enum values in double quotes. Use raw numbers and booleans.

For example:

• Find topics related to marketing: topic CONTAINS "marketing".
• Find high-performing mentions: avg_position <= 3 AND mention_count > 10.
• Find prompts with the enterprise tag: tags HAS "enterprise".