Combine tables that share columns

Stack files or workbook sheets that are one table split by period into a single table - and understand why a combine with no rows under the detected headers is refused.

4 min readUpdated November 2, 2026

Sometimes one dataset arrives split into pieces that all share the same columns — last month's rows in one file, this month's in another, or a workbook with one sheet per quarter. Combining stacks those tables into a single table, so you can chart across the whole thing at once instead of one slice at a time.

Note: Combining is for uploaded files and workbook sheets — data Golden holds locally. Tables from a live cloud connection (Snowflake, BigQuery, and the like) are queried in place, so the Combine action does not appear for them. To bring several related warehouse tables into one analysis, define how they join instead — see "Model several tables together".

When you upload

When the sheets in a workbook, or a batch of files you dropped in, look like one table split by period, Golden offers to combine them as they load. The dialog is titled "Sheets that look like the same table" (or "Files that look like the same table") and tells you how many pieces are in each set. Pick the set you want and they import as one table, with a column that records which sheet or file each row came from. Sheets that are genuinely different tables are left alone for you to pick individually.

Note: If the pieces you picked have no rows under the column names Golden detected — often because blank rows sit between the records — the combine is refused rather than producing an empty table: "No rows were found under the column names we detected in these sheets, so there is nothing to combine. If blank rows separate the records, remove them and try again." Clear the blank rows and upload again.

Re-adding a file that shares a table's name

If you add a file — or a workbook sheet — whose name matches a table the dataset already holds, Golden asks what to do rather than quietly adding a second copy. The dialog names the table it clashes with — for example "'Sales' is already a table here" — and offers three choices:

Replace the existing table — swaps the new file's rows in for the old ones. Charts and joins built on the table keep working, as long as the columns still match.
Add the rows to it — keeps what is already there and stacks the new rows on top, with a column recording which file each row came from. This is the same result as combining.
Add as a separate table — leaves the existing table alone and adds the new one alongside it under a new name.

Note: Replace is not offered for a table that already has rows stacked into it — that table is rebuilt from all its pieces every time the analysis loads, so a replace would not stick. Golden shows the reason in place of the option.

Combine tables already in your analysis

If your analysis already holds several tables, you can combine them from the data-model view — the diagram of your tables in Connect.

Open the data-model view, where each table shows as a card.
Click Combine tables in the toolbar (its tooltip reads "Stack two tables that hold the same columns into one"). You can also open a table card's menu and choose Combine With Another Table, or drag one card onto another.
In the Combine tables dialog, the "Keep this table" field is the table you're adding rows to, and "Add this table's rows into it" is the one being stacked in. Tables that line up are marked "looks like the same table".
Read the summary line, which spells out what will happen, then click Combine tables.

Note: The combined table keeps a column that records which source table each row came from. If one of the tables has an extra column the other doesn't, that column is added and left empty for the rows that never had it.

Note: The action only offers to stack tables that share enough columns. If two tables don't line up, Golden tells you so and suggests joining them on whatever they do share instead — combining is for two tables that are really one table split in two, not for related tables that reference each other.

Tip: A combined table's card shows how many pieces went into it (for example "3 sheets combined"). Click that strip to open the manage dialog, where you can add another matching sheet or remove one you no longer want.