Get SEO answers fast without opening Search Console. A simple chat input collects your question and returns data in a clear table. Marketing and SEO teams can check clicks, queries, and pages across many sites in minutes.

A secure webhook receives chatInput and sessionId. A set node adds a date hint for the agent. The AI agent uses an OpenAI chat model and Postgres memory to read your request, confirm the need, and choose the right tool. The tool calls a sub workflow that routes either to get the list of properties or to query custom insights. Set nodes build a request that follows the Search Console API. OAuth nodes pull the data. Array aggregation organizes the result for the agent. The final answer is sent back through the webhook.

You need OpenAI, Google Search Console access, and a Postgres database. Point the tool node to the tool trigger workflow, then activate and test with a POST request that includes chatInput and sessionId. Expect faster reports, fewer clicks, and clear answers to common SEO questions like top queries this week or pages losing clicks.

• Secure webhook with basic auth receives chatInput and sessionId
• Set node adds chatInput, sessionId, and a formatted date message
• AI agent uses an OpenAI chat model and Postgres chat memory with a context window of 5
• Tool workflow links the agent to a sub workflow when a tool call is needed
• Switch node routes between website list and custom insights paths
• Set nodes build the Search Console API body including dates, dimensions, and filters
• Google Search Console nodes fetch property lists and analytics data with OAuth 2
• Array aggregation prepares raw data for tables or charts
• Respond to webhook sends the final answer back to the caller

• Cut report build time from 60 minutes to 5 minutes
• Reduce manual Search Console clicks from 30 to 1 request
• Automate 80 percent of recurring SEO questions
• Improve data consistency by removing copy paste errors
• Support multiple sessions at once using sessionId
• Connect OpenAI, Search Console, and Postgres in one flow

1. Import the template into n8n: Create a new workflow in n8n > Click the three dots menu > Select 'Import from File' > Choose the downloaded JSON file.
2. You'll need accounts with OpenAI, PostgreSQL and Google Search Console. See the Tools Required section above for links to create accounts with these services.
3. Open the Webhook node named ChatInput. Confirm the HTTP method is POST and set Basic Auth credentials. Copy the production webhook URL for later testing.
4. Configure OpenAI: double click the OpenAI Chat Model node, choose the credential dropdown, click Create new credential, and paste your OpenAI API key from your OpenAI account. Select the model you prefer such as gpt 4o or gpt 4o mini.
5. Configure PostgreSQL: double click the Postgres Chat Memory node, create a new Postgres credential, and enter host, port, database, user, password, and SSL if needed. Set the table name to insights_chat_histories and keep context window at 5.
6. Connect Google Search Console: open either Search Console node, in Credential to connect with click Create new credential, then follow the on screen steps to sign in with the Google account that has Search Console access. Approve required scopes.
7. Set up the tool workflow link: open the Call Search Console Tool node and select the workflow that contains the Tool calling trigger. Make sure that workflow is saved and active.
8. Review Set fields nodes. Confirm chatInput and sessionId map from the webhook and that the date message format matches your time zone needs.
9. Open Set fields - Construct API CALL and verify request_type values website_list and custom_insights. Check that dates, dimensions, and filters map to your Search Console needs.
10. Test Google access: run the Get List of Properties node. If you see no sites, confirm Search Console permissions for your Google account and reauthenticate if needed.
11. End to end test: send a POST to the webhook with JSON that includes chatInput and sessionId. Example chatInput could be Show top queries for the last 7 days for site X. Verify you receive a table in the response.
12. Troubleshoot common issues: 401 or 403 from Google means credential or access problems, fix account permissions. Empty results often mean wrong date range or filters, adjust the Set fields node. Memory issues suggest database connectivity, check Postgres credentials and network rules.