Turn new Notion pages into search ready chunks that land in Pinecone within minutes. Teams can find fresh notes, docs, and briefs fast without manual copy and paste. Great for knowledge bases, content ops, and internal research.

A Notion event checks a chosen database every minute. When a page is added, the content is pulled and non text blocks are filtered out. The text is joined into one clean field, then split into 256 token chunks with a 30 token overlap to keep context between parts. Each page gets metadata like page ID and timestamps. Google Gemini creates embeddings from the chunks and Pinecone stores them in the selected index. This keeps your index fresh, clean, and easy to search with vectors.

You will need access to Notion, a Google Gemini API key, and a Pinecone index. Point the trigger at the right Notion database, confirm the chunk settings, and pick the model text embedding 004 in Gemini. Expect faster handoffs and less manual tagging. Common uses include indexing meeting notes, content drafts, product docs, and internal playbooks stored in Notion.

• Notion page added trigger polls the chosen database every minute for new pages.
• Retrieves all blocks from the page URL and collects full content.
• Filters out non text blocks so only readable text is processed.
• Concatenates block content into a single clean text field for reliable input.
• Splits text into 256 token chunks with 30 token overlap to preserve context.
• Adds metadata like pageId, createdTime, and lastEditedTime for better traceability.
• Generates embeddings using Google Gemini model text embedding 004.
• Inserts vectors into the Pinecone index for fast semantic search.

• Reduce manual copy and paste from 20 minutes to under 2 minutes per page
• Streamline content indexing by over 80 percent with automatic updates every minute
• Improve data quality by removing non text blocks before indexing
• Handle thousands of pages without changing the setup
• Connect Notion, Google Gemini, and Pinecone in one flow
• Keep search results fresh within one minute of page creation

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 Notion, Google Gemini and Pinecone. See the Tools Required section above for links to create accounts with these services.
3. In Notion, locate the target database and ensure your integration has access to it. Copy the database URL for reference.
4. In n8n, open the Notion - Page Added Trigger node. In the credential dropdown, click Create new credential and follow the on screen steps to connect your Notion workspace. Select the target database and keep the poll schedule at every minute.
5. Open the Notion - Retrieve Page Content node. Confirm Resource is Block and Operation is Get All, and that blockId uses the page URL from the trigger.
6. Open Filter Non Text Content and review the conditions so only text style blocks pass through. Save changes.
7. Open Summarize and confirm it concatenates the content field with a newline separator and outputs separate items.
8. Open Token Splitter and set chunk size to 256 and chunk overlap to 30 for balanced context and performance.
9. Open Embeddings Google Gemini. Click Create new credential, add your Google Gemini API key from the Google AI Studio API page, and select model text embedding 004.
10. In your Pinecone dashboard, create or select an index named notion-pages. Note the API key and environment.
11. Open the Pinecone Vector Store node. Click Create new credential, paste your Pinecone API key and environment, choose the notion-pages index, and keep Mode set to Insert.
12. Open Create metadata and load content and confirm jsonData uses the concatenated text expression and includes metadata fields like pageId, createdTime, and lastEditedTime.
13. Test the flow: add a new page in the Notion database with some text. Run the workflow once and check Pinecone for an increase in vector count.
14. If vectors do not appear, verify the Notion database ID, confirm the trigger has access, check Gemini API quotas, and make sure the Pinecone index name matches.
15. When tests pass, switch the workflow to active so new pages index automatically.