Sheets is your CRM, your log, your dashboard. Now your AI uses it too.
Most teams already have a spreadsheet running their CRM, their pipeline, their content calendar, their KPIs. Connect Google Sheets to Oya and your AI employee reads it, writes to it, appends rows, and renders reports — without leaving the workflow your team already trusts.
What it does
Read, append, and update
Pull data from named ranges or full sheets, append new rows (e.g. log every prospect contacted), and update specific cells by row/column or by query. Multi-sheet workbooks supported.
Search by content
Find rows that match a query ("all clients with status = active") without needing to know exact row numbers. Your agent can iterate over results and update each.
Reports and digests
Compile a weekly summary from a Sheet, format it as a Slack message, and post — all in one routine. Your agent can also write back the summary to a "Reports" sheet for record-keeping.
How to connect Google Sheets
- 1
Click Connect Google Sheets
On the agent’s Channels & Apps page, click Google Sheets, and pick the Google account that owns or has shared access to the spreadsheets you want.
- 2
Approve the Sheets scope
We request only Sheets API access — list, read, write. We don’t reach into Drive without a separate Drive connection.
- 3
Tell your agent which sheets it uses
In the system prompt or per-task instructions, name the spreadsheet(s) by ID or share URL. The agent treats those as the canonical source of truth for the workflows you assign.
- 4
Wire it into routines
Combine with Gmail (log incoming inquiries to a Sheet), with Slack (post weekly summaries from Sheets data), or with the Webhook gateway (append rows on form submissions).
Detailed walkthrough with screenshots: Channels & Apps in the docs.
What it doesn’t do (yet)
We’d rather be upfront about the edges than promise things we don’t deliver.
- Pivot tables and conditional formatting created in Sheets stay intact, but the agent doesn’t generate them — write your formulas in cells the agent populates instead.
- Apps Script triggers tied to Sheets are independent; if your sheet runs onEdit scripts, those still fire normally.
- Very large spreadsheets (50k+ rows) are supported but slower — for big datasets a real database makes more sense, and the agent can hit one over MCP or the API.
Frequently asked
Which spreadsheets can my agent access?
Any spreadsheet the connected Google account can open. Share a sheet with your agent’s connected Google account, and your agent can read or write to it. Don’t share, and the agent can’t see it.
Will the agent break my formulas or formatting?
No. The Sheets API is value-aware — the agent writes values into cells, leaving formulas, conditional formatting, and pivot tables alone. If you want the agent to fill a cell that should compute, point it at a different cell.
Can I have multiple agents writing to the same sheet?
Yes. Sheets handles concurrent writes well at small scale. For high-write workflows (multiple agents appending hundreds of rows per minute), watch for write contention — split into per-agent sheets if you see it.
Can I use Sheets as my CRM with this?
Plenty of teams do. The pattern: a Clients sheet (one row per client), a Pipeline sheet (one row per deal), and an Activity sheet (one row per contact). The agent reads from the first two to plan, writes to the third to log.
How do I disconnect?
From the agent’s Channels & Apps page, click Disconnect on the Sheets tile. Tokens are revoked immediately, and you can also revoke from your Google Account.
Does Oya store my spreadsheet data?
Only what the agent processed during a run, captured in your run history for audit. We don’t copy, sync, or index your sheets in the background, and we never use your data to train any model.
A note on AI accuracy
Oya’s AI employees are powered by large language models and may produce inaccurate, incomplete, or out-of-date responses. They are tools to assist your work, not authoritative sources. Always verify important information — especially numbers, dates, names, and any business or financial action — before relying on AI-generated output. If your AI employee suggests a destructive action (deleting data, sending money, posting publicly), confirm it yourself before approving.
No model training
Your data is not used to train AI models. It’s used to do the work you asked for, then logged in your run history for your audit.
Full audit trail
Every AI employee action maps to a run in Oya. Inputs, outputs, tools used, cost — all visible and auditable.
Your spreadsheet just learned to do the work itself.
Connect Google Sheets in one click. $10 in free credits to start.