Microsoft Excel 365 n8n integrationAutomate Microsoft Excel 365 with n8n.

Rows land at the bottom of an existing table object, and the table keeps its own column definitions.

• Workbook, Sheet and Table: three pickers, each filtered by the choice above it.
• Data: the raw values as an array of string arrays in JSON, one inner array per row.
• Data Mode: autoMap when the incoming item carries the column names, define to set each column by hand, raw to send the JSON untouched.
• Index: left empty, the row goes last; a position slots it in and pushes the rows below down, first row index 0.

The table object goes away while the cells stay where they are, which turns a structured table back into ordinary cells.

• Workbook and Sheet: the file and the tab holding the table.
• Table: the table to unwrap, picked from the list or given as an ID through an expression.

A plain range becomes a table, which is what the other table operations need before they can do anything.

• Workbook and Sheet: where the new table gets created.
• Select Range: auto converts the whole used range of the sheet, manual hands the choice to the next field.
• Range: the cells to convert, in A1 style such as A1:B2.
• Has Headers: off, Excel generates headers itself and pushes the data down by one row.

One call removes the table from the sheet. Convert to range is the other way out when the values have to stay.

• Workbook, Sheet, Table: the same three step path as every table operation, down to the named table.

The structure comes back rather than the content: the list of columns the table declares.

• Workbook, Sheet, Table: the table to inspect.
• Return All: on, every column comes back; off, Limit caps the count.
• Fields: a comma separated list to trim the response down to what the next node reads.

Each row of the table arrives as its own item, keyed by the header names, ready for a loop or a filter.

• Workbook, Sheet, Table: the source table.
• Return All: leave it off and the node stops at Limit, which is the usual cause of an export that looks truncated.
• Column Names or IDs: the subset of columns to bring back.
• RAW Data: on, the response is not parsed into keys, and Data Property names the field it lands in.

Instead of pulling the table and filtering afterwards, this one searches a single column and returns the rows that match.

• Lookup Column: the column name to search in, Email for instance.
• Lookup Value: what to find there, often {{ $json.email }} coming from the node before.
• Return All Matches: off by default, so only the first hit comes back; switch it on when duplicates matter.

A new tab appears in the workbook, empty and ready to be written into by a later node.

• Workbook: the file receiving the sheet, from the list or through an expression.
• Name: optional but worth setting, since Excel names the sheet itself when the field stays empty; the name has to be unique in the workbook.

The whole file goes, not a tab and not a range, so this one belongs behind a condition rather than in the main path.

• Workbook: the file to delete, chosen from the list or resolved by ID through an expression.

No picker to fill here: the operation lists the workbooks the credential can see, with their identifiers.

• Return All: on for the full inventory, off to stop at Limit.
• Fields: a comma separated list, useful when only the name and the ID matter downstream.

Values go straight under the used range of a sheet, with no table object involved anywhere.

• Workbook and Sheet: the file and the tab that receives the rows.
• Data: an array of string arrays in JSON, one inner array per row to write.
• Data Mode: autoMap maps input properties onto matching column names, define exposes Values to Send to build each column, raw passes the JSON along as it is.

This one reads before it writes: an existing row gets updated, a missing one gets added, in a single node.

• Column to match on and Value of Column to Match On: the pair that identifies the row; the matching column is never modified.
• Select a Range and Range: a specific A1 style range whose first row holds the column names, or blank to use the whole used range.
• Update All Matches: on, every duplicate is rewritten, not only the first one.

The sheet stays in place while what it holds goes away, entirely or over a chosen range.

• Apply To: Contents wipes the data, Formats wipes the font sizes and colors, All takes both.
• Select a Range and Range: an A1 style range to limit the sweep; left blank, the whole worksheet is cleared.

Here the tab itself disappears from the workbook, which is the difference with Clear sheet.

• Workbook: the file to work in.
• Sheet: the tab to remove, picked from the list or given as an ID through an expression.

A workbook rarely holds a single tab, and this operation returns the list of sheets it contains.

• Workbook: the file to open.
• Return All: on, every sheet comes back; off, Limit decides how many.
• Fields: a comma separated list to keep the response small.

Reading works on a sheet that was never turned into a table, which is the case of most files people send around.

• Range: A1 style and specific, A1:B5 is fine and A:B is not; blank returns the entire sheet.
• Header Row: the index of the row holding the column names, counted inside the range, first row index being 0.
• First Data Row: the index where the real rows start, which saves a sheet with a title and a blank line above the table.

Existing rows get overwritten with new values, either across a range or on the rows that match a column.

• Data: an array of string arrays that has to match the range, one inner array per row.
• Column to match on and Value of Column to Match On: how the node finds the right row when the update is targeted.
• Data Mode: raw rewrites the whole selected range with the new values, autoMap and define work column by column.
• Update All Matches: on to rewrite every matching row.