The Scenario
Every Monday morning you pull five Metabase saved questions — DAU, conversion rate, churn, LTV, ARR — into a single Google Sheet before the leadership meeting at 9 AM. You've done this eight Mondays in a row. Each time: open Metabase, navigate to the first card, run it, download the CSV, switch tabs, paste. Then repeat for the next four.
The bad version:
- Open each of the five saved questions in separate Metabase tabs, run them in sequence, and download a CSV for each.
- Import or paste each CSV into the correct tab of the Google Sheet, in the correct order, being careful not to overwrite the formula rows at the bottom.
- Delete the five CSV files from Downloads before your laptop storage warning reappears.
Last week card 103 — the LTV question — had been edited overnight. The column order changed. You didn't notice until leadership pointed out that the LTV column showed conversion rate numbers. You re-did the paste at 9:04 AM with the meeting already in progress.
The Easy Way: One Prompt in SheetXAI
SheetXAI is an AI agent that lives inside your Google Sheet. It can execute multiple Metabase saved question cards in a single prompt and write each result into a separate tab, named and ordered the way you specify.
Run Metabase card queries for card IDs 101, 102, 103, 104, and 105 and write each result into a separate tab named after the card — include column headers on each tab
What You Get
- Five tabs are created (or overwritten if they already exist), each named after the corresponding card.
- Each tab has a header row matching the columns the saved question returns.
- The data reflects the latest query execution — not a cached result from the last time someone viewed the card in Metabase.
- If a card returns zero rows (say, no churned users this week), the tab gets the header row and a note rather than a blank sheet that looks like a write failure.
What If the Data Is Not Quite Ready
One of the card IDs has changed since you last ran this
For each Metabase card ID in column A of the Index tab, execute the card query and write the first result row's key values into columns B through F on the same row, then write the full result into a separate tab named for the card
Driving the card list from the sheet means you update the list once — in the sheet — rather than editing the prompt every time a card gets archived and recreated with a new ID.
You need results stacked vertically instead of across tabs
Run Metabase card queries for card IDs 101, 102, 103, 104, and 105 and write each result into Sheet1, stacking them vertically with the card name as a section header and a blank row between each
A single flat tab, section-labeled, ready for a weekly email paste or a PDF export.
One card's result needs to be filtered before it lands in the sheet
Run Metabase card 102 and write only the rows where conversion_rate is greater than 0.05 into the Conversion tab with column headers
The filter applies before the data reaches the sheet — you're not loading 800 rows and hiding the rest with a view filter.
Full refresh plus data quality check in one shot
Run Metabase card queries for card IDs 101, 102, 103, 104, and 105, write each into its own tab with headers, then in a Summary tab write one row per card showing the card name, row count, and whether any numeric column contains a null value
One prompt: refresh five cards, write the tabs, and produce a data quality summary. Leadership gets the numbers; you get a signal if anything is worth investigating before you walk into the meeting room.
The pattern: tell SheetXAI what you need across all five cards in one go. It handles the sequencing.
Try It
Get the 7-day free trial of SheetXAI and open the Google Sheet you use for your weekly leadership metrics, then ask it to pull all your Metabase saved questions into their tabs in one pass. Also useful: running a custom SQL query from Metabase into a sheet, and the hub overview on connecting Metabase to Google Sheets.