How to Connect Airtable and Google Sheets: Sync, Export, and What Breaks

I went looking for a straight answer to this and could not find one. The first page of results is Airtable’s own forum, Airtable’s own documentation, and about ten companies that sell a connector. The forum threads are people asking each other. The vendor pages all end at the same place, which is their product.

So here is the version I wanted: what each method actually does, what breaks, and the question worth answering before you connect anything.

I build operations systems on Airtable for a living, and I use Google Sheets alongside it every week. I do not sell a connector.

The three methods, and what each one can actually do

The second row is the one that catches almost everybody, so it gets its own section.

Can Airtable’s automation update an existing row in Google Sheets?

No. It can only add a new row to the bottom.

This is the single most important thing on this page, and it is stated almost nowhere. Airtable’s native automation has an action that appends a row to a Google Sheet, and it has no action that finds an existing row and changes it.

So the behaviour people report goes like this: you set up an automation on “when a record is updated”, you change a value in Airtable, and instead of your Google Sheet updating, a duplicate row appears at the bottom. The automation did exactly what it was built to do. It just was not built to do what you assumed.

The same applies in reverse. A new row in Google Sheets can create a record in Airtable. A change to an existing row cannot update the matching record.

That gap is the reason the connector market exists. Every tool selling you an Airtable/Sheets sync is selling you the update operation, whatever else the marketing says.

How can I integrate Airtable with Google Sheets?

Manual export and import. Download a CSV from an Airtable view, then use File → Import in Sheets. Fine for a one-off. It is a snapshot, so it is stale the moment anyone touches either side, and complex Airtable fields (attachments, linked records, checkboxes) rarely survive the trip intact.

Airtable’s native automation. In your base, Automations → Create new, pick a trigger, then the Google Sheets action, then map fields to columns. Genuinely useful for append-only jobs: a log, an audit trail, a running list of submissions. Not useful for keeping two views of the same data in agreement, for the reason above.

A third-party connector. Zapier, Make, n8n, and a set of dedicated sync tools all do this, and they are the only option that updates rather than appends. Setup is field mapping plus a schedule, and the thing to get right is the matching key: the connector needs a stable unique field on both sides to know that this row and that record are the same thing. Without one it creates duplicates, which is the problem you were trying to solve.

Can you sync in both directions?

You can, and it is harder than any vendor page makes it sound. The reason is worth understanding, because it is not a limitation any tool can fix.

Google Sheets is not a database. It has no transaction control, no rollback, and no conflict management. If a person edits a cell at the same moment a sync writes to it, there is no mechanism that decides who wins. Every integration platform hits this same wall, because the limitation lives in Sheets, not in the connector.

In practice that means two-way sync works when edits on the two sides are naturally separated, by time, by column, or by who does them. It gets unreliable when the same field is genuinely being edited in both places, and no amount of configuration fixes that. It is a data-ownership problem wearing an integration problem’s clothes.

Why did my sync stop working?

Because the Google authorization expired, most likely. This is the most common failure and almost nobody writes about it.

The symptom is an Airtable automation that ran fine for months and then stopped, with an error along the lines of this automation is no longer authorized to use the configured Google Sheets account. Reconnecting the account usually fixes it.

The part that costs you is not the error. It is that nothing tells you. The automation quietly stops, the Google Sheet keeps showing yesterday’s numbers, and someone finds out in a meeting a fortnight later.

If a sync matters, build something that notices when it dies. A scheduled check that compares record counts on both sides, or a timestamp field that a human can glance at, is enough. Any signal beats finding out from a client.

How do I import a Google Sheet into Airtable?

From your base, + Add or import → Google Sheets, connect the Google account, choose the sheet, and adjust field types in the preview before confirming. You can import as a new table, into a new base, or into an existing table.

Two limits worth knowing before you plan around it:

Size. Native import caps at 100MB for CSV and 5MB for Excel. Past that you split the data across bases and sync them together.

Bulk. You cannot import multiple spreadsheets at once, natively or through the CSV extension. If you have fifty sheets with the same structure, you either merge them first or build an automation in Make or n8n to loop the import. This one is worth checking before you commit to a plan, because people discover it after buying.

And a distinction that catches people: importing is not connecting. Once the import finishes, the two are unrelated. Changes in the Google Sheet do not reach Airtable and changes in Airtable do not reach the sheet.

How do I export Airtable to Google Sheets?

Airtable export is the one-directional case, and it is a different job from connecting. Airtable sends the data out to a sheet once, or on a schedule, and nothing comes back.

One view at a time, by hand. Open the view, right-click it in the Views menu, Download CSV, then in Sheets use File → Import. Airtable exports the view rather than the table, so whatever filters and hidden fields sit on that view are what lands in the sheet. That is useful when you want it and surprising when you forget it.

The whole base at once. Airtable can export a base as CSV files, one per table. You then import each file separately, because Sheets will not take them as a set.

On a schedule. Airtable has no native scheduled export. An automation in Make or n8n can pull an Airtable view and append it to a sheet on a timer, and connector tools sell the same thing as a product. Both work. The trade is the one on the whole page: something outside Airtable now has to keep running.

Three things that catch people:

• Airtable does not export attachments. It exports the file URL as text, and that link expires. If the images matter, export them separately.
• Airtable exports linked records as the primary field’s text, so a link to “Acme Corp” arrives as the words Acme Corp and stops being a relationship.
• An Airtable export is a snapshot. It is stale the second anyone edits either side, which is the whole reason people go looking for a sync instead.

If you find yourself exporting the same view every week, the export is not the answer you were looking for. Read the last section of this page before you automate it.

What breaks when the data moves across?

Field conversion is lossy in specific, documented ways. Airtable publishes the rules and nobody translates them into what will actually go wrong, so:

• Currency only recognises US dollars. Other currency symbols do not convert to a currency field.
• Time-only values become text. A cell containing 18:02 with no date lands as single-line text.
• Checkbox conventions cannot be mixed. A column with both yes and x converts to text rather than checkboxes.
• Multi-select conversion is fussy. It fails on repeated tokens in one cell, punctuation other than commas, tokens over 20 characters, or tokens that are only digits.
• Formulas convert their result, not the formula. You get the number that was showing, and the logic behind it is gone.
• Dates get a format inferred from your data. Mixed formats in one column produce ISO, which may not be what anyone expected.

None of this is a bug. It is what happens when a structured system reads an unstructured one, and it is exactly why the cleanup, not the import, is the actual work in any migration. That is also true of the larger move, which is covered in Airtable vs Google Sheets.

Should you be syncing at all?

Every page I read while researching this assumes the answer is yes and competes on how. It is worth asking anyway.

A sync is the right answer when Sheets is a read-only surface. Airtable owns the data, Sheets is where someone builds a pivot table, models a scenario, or hands a client a tab they can filter. Data flows one direction, nobody edits the Sheets copy, and the thing stays boring for years. That setup works and I build it often.

A sync is the wrong answer when both sides are being edited. At that point you do not have an integration problem. You have two systems that each believe they are the source of truth, and a connector will not resolve that. It will just move the disagreement around on a schedule.

The question to answer before you connect anything is which system owns each piece of data. Sometimes the honest answer is that the work living in Sheets should not be in Sheets, and the sync you were about to build is a way of avoiding that conversation. The same question applies one step earlier, at the point of collection, where a Google Form and an Airtable form send the data to very different places. Our Airtable consulting work usually starts by drawing that line, and the operations side covers what to do when the line runs through a process rather than a tool.

Frequently asked questions

Does Airtable have a native Google Sheets integration?

Yes, through automations, and it is narrower than the name suggests. It can create a row in Google Sheets from an Airtable record and create an Airtable record from a new Google Sheets row. It cannot update anything that already exists on either side.

Can I sync Airtable and Google Sheets for free?

For append-only flows, yes, using Airtable’s native automations on any plan. For anything that updates existing rows you need a third-party connector, and scheduling is usually the feature that sits behind the paid tier.

What is the best tool to connect Airtable and Google Sheets?

It depends on what else you are connecting. If this is the only integration you need, a dedicated sync tool is simplest. If you already run automations elsewhere, use what you have: Make and n8n both handle it and you avoid another subscription and another thing to monitor.

Why is my Airtable automation creating duplicate rows in Google Sheets?

Because the append action is the only one it has. Every trigger adds a new row rather than finding and updating the matching one. To update rather than append, you need a connector with a matching key.

Is there a way to export Airtable data?

Yes, several. Airtable exports a view as CSV, a whole base as one CSV per table, or on a schedule through an automation. Full detail is in the export section above. Airtable does not carry attachments or linked records through intact in any of them.

Can Google Sheets replace Airtable?

For a flat list that one or two people maintain, easily. It stops being possible once one record needs to relate to another, once several people edit at the same time, or once submissions need to become work on their own. That is a different question from this page, and it has its own answer in Airtable vs Google Sheets.

Can Airtable be the source of truth with Google Sheets as a view?

Yes, and it is the setup most likely to keep working. Airtable holds and validates the data, a scheduled one-way push keeps a Sheets copy current, and everyone agrees that edits made in the sheet get overwritten. The agreement matters more than the tooling.

If you are weighing whether Airtable should hold the data in the first place, a plain description of Airtable covers what it is and where it stops.