Google Sheets

Read and write to Google Sheets from your agent.

Read from and write to Google Sheets for lookup tables, call logging, or dynamic content.

Quick setup (Paragon integration)

Connect Google Sheets directly from the Integrations page in Agent Studio using OAuth.

1. Go to the Integrations page in Agent Studio
2. Find Google Sheets and click Connect
3. Select an environment (Sandbox, Pre-release, or Live)
4. Click Connect new account and authenticate with Google via OAuth
5. Enable Google Sheets actions in your agent’s prompt

Available actions

Action	Description
Get Row	Read row data from a Google Sheets spreadsheet for reference

Enabling actions

Reference Google Sheets actions in your agent’s prompts. See the Zoom integration for detailed examples of this pattern.

---

Manual setup (service account)

For advanced use cases requiring write access or custom API calls, you can integrate Google Sheets manually with a service account.

Capabilities

• Read data: Query spreadsheet cells, rows, or ranges
• Write data: Append new rows or update existing cells
• Search: Find specific values within a spreadsheet

Getting started

Prerequisites

• A Google account with access to Google Sheets
• The spreadsheet(s) you want to connect
• PolyAI project access

Step 1: Prepare your spreadsheet

1. Create or open the Google Sheet you want to use
2. Ensure the first row contains clear column headers
3. Note the Spreadsheet ID from the URL:
   https://docs.google.com/spreadsheets/d/{SPREADSHEET_ID}/edit
   

Step 2: Create a service account

1. Go to the Google Cloud Console
2. Create a new project or select an existing one
3. Enable the Google Sheets API
4. Navigate to APIs & Services → Credentials
5. Click Create Credentials → Service Account
6. Fill in the service account details and create
7. Click on the service account, then Keys → Add Key → Create new key
8. Select JSON and download the key file

Step 3: Share the spreadsheet

1. Open your Google Sheet
2. Click Share
3. Add the service account email (found in your JSON key file) as an editor
4. Click Send or Share

Step 4: Provide credentials to PolyAI

Share the following with your PolyAI team:

• Service account JSON key file (securely transferred)
• Spreadsheet ID(s)
• Sheet names and structure description

PolyAI will store your credentials securely and configure the integration.

Use cases

Lookup tables

Store reference data that your agent can query:

Product Code	Name	Price	Availability
SKU-001	Widget A	$19.99	In Stock
SKU-002	Widget B	$29.99	Limited

Call logging

Append call data for tracking:

Timestamp	Caller Number	Intent	Resolution
2024-01-15 10:30	[phone]	Booking inquiry	Completed

Dynamic content

Store content that changes frequently:

Topic	Response
Holiday hours	We’re closed Dec 25-26
Special offer	20% off all bookings this week

Limitations

• Rate limits: Google Sheets API has usage quotas; high-volume operations may require optimization
• Data size: Best suited for small to medium datasets; large datasets should use a proper database
• Concurrent access: Multiple simultaneous writes may cause conflicts; design your sheet structure accordingly
• Latency: API calls add slight delay compared to cached data

Best practices

1. Keep data clean: Use consistent formatting and avoid merged cells
2. Use named ranges: Makes queries more reliable than cell references
3. Limit sheet size: Archive old data to maintain performance
4. Test thoroughly: Verify read/write operations before going live

Support

For setup assistance, contact your PolyAI account manager.