The agent uses Google Sheets' values.get and lookup APIs to search by specific column values. When a customer provides an email or order number, the agent scans the designated column, finds the matching row, and returns the relevant cell values. It works like VLOOKUP but runs during a live conversation.

Yes. The agent uses the spreadsheets.values.append endpoint to add new rows and values.update to modify existing cells. Whether it is logging a new lead or updating a customer's status field, the write happens in real time and appears in your sheet immediately.

Tars requests the spreadsheets OAuth scope, which covers reading and writing data across your Google Sheets. You authorize through one-click OAuth and can revoke access from your Google Account settings. The agent only accesses spreadsheets you explicitly configure in the workflow.

No. Tars reads from and writes to your Google Sheets in real time through the API. Cell values, formulas, and metadata are fetched live during conversations and are not cached or stored on our servers. Your spreadsheet data stays in Google's infrastructure.

Yes. The agent uses targeted range queries and lookup functions rather than loading entire sheets into memory. For large datasets, it searches specific columns efficiently. Google Sheets API supports spreadsheets with millions of cells, and the agent's query approach keeps response times fast even for big sheets.

Google Sheets handles concurrent writes at the platform level. Each append operation adds a new row atomically, so simultaneous conversations do not overwrite each other. For cell updates, the agent targets specific row and column coordinates, minimizing conflict potential.

The agent can read the computed values of formula cells, returning the calculated result rather than the formula string. When writing data, the agent can insert values that existing formulas reference, causing automatic recalculation. It does not modify formulas themselves unless explicitly configured to do so.

Google Forms collects data through a rigid, pre-defined form layout. Tars AI agents have conversations, asking follow-up questions, providing answers, and adapting the flow based on customer responses. The data still lands in Google Sheets, but the collection experience is dynamic and personalized instead of static and one-directional.